New Economic Districts in Belgrade Area more

NEW ASPECTS OF URBAN PLANNING AND TRANSPORTATION Proceedings of the WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07) Heraklion, Crete Island, Greece, July 22-24, 2008 Mathematics and Computers in Science Engineering A Series of Reference Books and Textbooks Published by WSEAS Press www.wseas.org ISBN: 978-960-6766-87-9 ISSN: 1790-2769 NEW ASPECTS OF URBAN PLANNING AND TRANSPORTATION Proceedings of the WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07) Heraklion, Crete Island, Greece, July 22-24, 2008 Mathematics and Computers in Science Engineering A Series of Reference Books and Textbooks Published by WSEAS Press www.wseas.org Copyright © 2008, by WSEAS Press All the copyright of the present book belongs to the World Scientific and Engineering Academy and Society Press. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior written permission of the Editor of World Scientific and Engineering Academy and Society Press. All papers of the present volume were peer reviewed by two independent reviewers. Acceptance was granted when both reviewers' recommendations were positive. See also: http://www.worldses.org/review/index.html ISBN: 978-960-6766-87-9 ISSN: 1790-2769 World Scientific and Engineering Academy and Society NEW ASPECTS OF URBAN PLANNING AND TRANSPORTATION Proceedings of the WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07) Heraklion, Crete Island, Greece, July 22-24, 2008 Editors: Prof. Thomas Panagopoulos, University of Algarve, PORTUGAL Prof. Jon Bryan Burley, Michigan State University, USA Prof. Selma Celikyay, Zonguldak Karaelmas University, TURKEY International Program Committee Members: Laszlo Garbai, HUNGARY Dorde Kozic, SERBIA and MONTENEGRO Darko Goricanec, SLOVENIA A. C. Benim, Germany Abul-Fazal M. Arif, SAUDI ARABIA Agis Papadopoulos, GREECE Ahmed Hassan, EGYPT Ahmed Mohammadein, EGYPT Alexander Kuzmin, RUSSIA Ali J. Chamkha, KUWAIT Aly Elshamy, EGYPT Ana Sirviente, USA Andrei G. Fedorov, USA Aroudam El hassan, MAROCCO Asad Salem, USA Aura L. Lopez de Ramos, VENEZUELA Aydin Misirlioglu, TURKEY Beghidja Abdelhadi, FRANCE Bodo Ruck, GERMANY Boris Ushakov, RUSSIA Bouhadef Khedidja, ALGERIA Bouhadef Malek, ALGERIA Bozidar, CROATIA, C. Treviño, MEXICO C.W. Leung, HONG KONG Chang Kyun Choi, KOREA Claudia del Carmen Gutierrez-Torres, MEXICO David Katoshevski, ISRAEL Domenico Guida, ITALY Dragoljub Mirjanic, BOSNIA AND HERZEGOVINA Federico Mendez, MEXICO Fereydoun Sabetghadam, IRAN Florin Popescu, ROMANIA Fotis Sotiropoulos, USA Francoise Daumas-Bataille, FRANCE Gareth Thomas, USA Gennaro Cardone, ITALY Günter K.F., GERMANY H.S. Takhar, UK Hany Mohamed, EGYPT Haris Catrakis, USA Henar Herrero, SPAIN Hossein Shokouhmand, IRAN Hyung Hee Cho, KOREA Ivan Kazachkov, SWEDEN Jean-Christophe Robinet, FRANCE Jeong-se Suh Gyeongsang, KOREA Jing Liu, CHINA Joakim Wren, SWEDEN Joseph T. C. Liu, USA Junjie Gu, CANADA K. P. Sandeep, USA Kadir Bilen, TURKEY Kai H. Luo, UK Khaled Alhussan, SAUDI ARABIA Krish Thiagarajan, AUSTRALIA Luis Cortez, BRAZIL M. Abu-Zaid, JORDAN Mahmoud Jamiolahmady, UK Md Anwar, BANGLADESH Mehdi Azhdary Moghaddam, IRAN Mehmet C. Ece, TURKEY Michiharu Narazaki, JAPAN Mohd Al-Nimr, JORDAN Mostafa Mahmoud, EGYPT Muthukumaran Packirisamy, CANADA Nabil Moussa, EGYPT Nicolas Galanis, CANADA Nikolaos Markatos, GREECE Oh-hyun Rho, Seoul National University, KOREA Prof. Oleg V. Vasilyev, USA Olga Mazhorova, RUSSIA Omar Abdel-hafez, EGYPT P V S N Murthy, INDIA Pablo S. Casas, SPAIN Pascal Roubides, USA Pavel Krukovsky, UKRAINE Pradip Majumdar, USA Pradipta Panigrahi, GERMANY Rafael Royo, SPAIN Ramil Sharafutdinov, RUSSIA Roger Grimshaw, UK Ryszard Tadeusiewicz, POLAND Serkan Ozgen, TURKEY Shabaan Abdallah, USA Shoaib Usman, USA Siavash Sohrab, USA Slawomir Smolen, GERMANY Somchai Wongwises, THAILAND Sujoy Kumar, INDIA Suman Chakraborty, INDIA Tahira Haroon, PAKISTAN Tamas Reti, HUNGARY Tatsuo Inoue, JAPAN Valeri Bubnovich, CHILE Viorel Stoian, ROMANIA Vitoriano Ruas, FRANCE Yinping Zhang, CHINA Yizhen Huang, CHINA Yue Dong, CHINA Andrej Krope, SLOVENIA Nicolas Abatzoglou, CANADA Beghidja Abdelhadi, FRANCE Dimitris Achilias, GREECE Wael Al-hasawi, KUWAIT Zakaria Al-Qodah, JORDAN Tomas Bodnar, CZECH REPUBLIC Luis Borges, PORTUGAL Corneliu Botan, ROMANIA Arturo Bretas, BRAZIL Fernando Carapau, PORTUGAL Sombat Chuenchooklin, THAILAND Paulo Correia, PORTUGAL Abdel-Karim Daud, ISRAEL Paul Deuring, FRANCE Tsanakas Dimitrios, GREECE Nikolay Djagarov, BULGARIA Jassim Gaeb, JORDAN Paschalis Grammenoudis, GERMANY Mohamed Hassan, KUWAIT Iraj Hassanzadeh, IRAN Toshiaki Hishida, JAPAN Seied Hossein, Hosseiny IRAN Chun Chang Huang, CHINA Niranjan Kumar Injeti, INDIA Lucio Ippolito, ITALY J. Janela, PORTUGAL Sameer Khader, ISRAEL Stanislav Krasmar, CZECH REPUBLIC Rainer Krebs, GERMANY Petr Kucera, CZECH REPUBLIC Sonia Leva, ITALY Bugaru Mihai, ROMANIA Ebrahim Mussavi, IRAN Jiri Neustupa, CZECH REPUBLIC Juan Ospina, COLOMBIA Hassan Rahimzadeh, IRAN Dong-Hee Rhie, KOREA Nasreddine Saadouli, KUWAIT B. Saf, TURKEY Maria Specovius-Neugebauer, GERMANY Frank Stagnitti, AUSTRALIA Mladen Stanojevic, SERBIA and MONTENEGRO Heiki Tammoja, ESTONIA Juhan Valtin, ESTONIA Werner Varnhorn, GERMANY George Verros, GREECE Mohamed Zahran, EGYPT Jiri Zdenek, CZECH REPUBLIC Gaetano Zizzo, ITALY Juan Zolezzi Cid, CHILE Preface This book contains the proceedings of the WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07) which was held in Heraklion, Crete Island, Greece, on July 22-24, 2008. This conference aims to disseminate the latest research and applications in Urban Planning Strategy and Development, Urban Planning Management, Environmental Protection, Financial Topics on Urban Planning, Architectural Heritage, Architectural Tradition and Development,Urban Tourism, Transportation Systems and Environment, Transportation Engineering, Monitoring of Transportation Systems, Sustainable Development and Transportation, Tourism and Transportation, Public Health and Transportation, Influence of Transportation Systems on Real Estate and other relevant topics and applications. The friendliness and openness of the WSEAS conferences, adds to their ability to grow by constantly attracting young researchers. The WSEAS Conferences attract a large number of wellestablished and leading researchers in various areas of Science and Engineering as you can see from http://www.wseas.org/reports. Your feedback encourages the society to go ahead as you can see in http://www.worldses.org/feedback.htm The contents of this Book are also published in the CD-ROM Proceedings of the Conference. Both will be sent to the WSEAS collaborating indices after the conference: www.worldses.org/indexes In addition, papers of this book are permanently available to all the scientific community via the WSEAS E-Library. Expanded and enhanced versions of papers published in this conference proceedings are also going to be considered for possible publication in one of the WSEAS journals that participate in the major International Scientific Indices (Elsevier, Scopus, EI, ACM, Compendex, INSPEC, CSA .... see: www.worldses.org/indexes) these papers must be of high-quality (break-through work) and a new round of a very strict review will follow. (No additional fee will be required for the publication of the extended version in a journal). WSEAS has also collaboration with several other international publishers and all these excellent papers of this volume could be further improved, could be extended and could be enhanced for possible additional evaluation in one of the editions of these international publishers. Finally, we cordially thank all the people of WSEAS for their efforts to maintain the high scientific level of conferences, proceedings and journals. WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 Table of Contents Plenary Lecture I: 100 Years of Highway Transportation Research Manoj K. Jha 9 An urban designers’ road hierarchy Taeke M. De Jong 11 Urban design transformations reducing public paved surface Taeke M. De Jong 17 Developing a Mobile GIS for Field Geospatial Data Acquisition Ebrahim Poorazizi, Ali A. Alesheikh 23 Developing a Genetic Algorithm for Solving Shortest Path Problem Saeed Behzadi, And Ali A. Alesheikh 28 New Economic Districts in Belgrade Area Slavka Zekovic, Tamara Maricic 33 Sustainable Transportation Planning:A Methodological Approach Jorge Diaz 39 Environmental Quality of Roads in Heraklion, Crete Androniki Tsouchlaraki, Eleftheria Zoaki 47 Theoretical Aspects Concerning the Adoption of the Number and Dimensions of Brake Discs for 56 Railway Vehicles Cruceanu Catalin, Craciun Camil Evaluation of the Sustainable Ecological Pond Management in Campus, Miaoli, Taiwan Kuei-Yang Wu, Wann-Ming Wey, Chi Ou-Yang, and Ya-Chung Lin 62 Non-Linear Multi-Objective Transportation Problem : A Fuzzy Goal Programming Approach Hamid Reza Maleki Sara Khodaparasti 69 Drivers stated preferences towards road pricing: the case of Athens Christina Milioti, Ioanna Spyropoulou, Matthew Karlaftis 74 Analyzing the types of urban physical development deviation using the structure of PERT technique, a glance to biology Wahid Yeganegi Dastgerdi 83 Customer Satisfaction of Light Rail Transit Sytem in Malaysia Ng Ying Peng, Wan Rosmanira Ismail, Nur Riza Suradi, Zainol Mustafa, Zalina Mohd Ali, Rofizah Mohammad ISBN: 978-960-6766-87-9 7 87 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 & Faridatulazna Ahmad Complex network applications to the infrastructure systems: the Italian Airport Network case Joseph Quartieri, Michele Guida, Claudio Guarnaccia, Salvatore D’ambrosio, Davide Guadagnuolo 96 Importance of Phase Changes in Balanced Growth of a City Wahid Yeganegi Dastgerdi 101 A Modified Arc Routing Problem for Highway Feature Inspection Considering Work-Shift and Overtime Limit Constraints Manoj Jha, Francis Udenta, Simon Chacha, Gautham Karri 105 Reconfiguring the Existing US Road System for Sustainable Living Diane Jones, Manoj K. Jha 110 Adaptation of New Dwelling Houses in East Towns to the Historical Environment (Case of Iran and Azerbaijan Republic) Javad Abdolhoseyni 116 Landscape Change in a Mountainous Area in Northeastern Portugal: Implications for Management César Moreira, João P. Castro, João Azevedo 122 Author Index 127 ISBN: 978-960-6766-87-9 8 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 Plenary Lecture I: 100 Years of Highway Transportation Research Dr. Manoj K. Jha Associate Professor and Director Center for Advanced Transportation and Infrastructure Engineering Research Department of Civil Engineering,Morgan State University,1700 East Cold Spring Lane Baltimore, MD 21251 U.S.A. Phone: (443) 885-1446, Fax: (443) 885-8218 E-mail: mkjha@eng.morgan.edu Web: http://www.eng.morgan.edu/~mkjha Abstract: Since the advent of automobile, and design and construction of paved roads in the early 1900’s there has been a significant contribution made in highway related transportation research. While in the earlier decades bulk of highway transportation research revolved around planning, design and construction of roads, right-of-way acquisition, and highway financing issues, the focus has somewhat shifted in the last three decades. Recent research has revolved around issues related to traffic congestion and environmental impact. This presentation reviews the changing trend in highway transportation research over the last 100 years and presents some views for future research. Specifically, it emphasizes the need for a closer examination of the three-way interaction among vehicle, roadway, and driver for better planning, design, and traffic management. There is a considerable trade-off among these three entities when planning, designing, operating, and maintaining highways as well as highway traffic. The presentation concludes with some discussion on effects of globalization and sustainability in highway transportation research. Brief Biography of the Speaker: Dr. Manoj K. Jha is Associate Professor of Civil Engineering at the Morgan State University where he oversees the transportation engineering graduate program and also serves as the Director of the Center for Advanced Transportation and Infrastructure Engineering Research. He obtained a Ph.D. in Civil Engineering with transportation specialization from the University of Maryland, College Park in 2000; a M.S. degree in Mechanical Engineering from the Old Dominion University in 1993; and a B.E. degree in Mechanical Engineering from the National Institute of Technology, Durgapur, India in 1991. He also attended the Rensselaer Polytechnic Institute during 1993-94 as a Ph.D. student in Mechanical Engineering and Virginia Tech,’s National Capital campus as a post doctoral fellow during 2000-2001. Dr. Jha’s research interests are in Highway Design Optimization and Visualization, Highway Maintenance, Geographic Information Systems, and Bayesian Network Application for assessing impacts of extreme events under uncertainty. For his scholastic and research achievements Dr. Jha has received several awards, among which are the 2005 and 2006 United Negro College Funds Special Program/Department of Defense (UNCFSP/DoD) Faculty Development Award, 2005 Department of Homeland Security (DHS) Summer Faculty Research award by the Study of Terrorism and Responses to Terrorism (START) Center of Excellence, University of Maryland, College Park, and 2005 NSF-PASI-TS (National Science Foundation’s Pan-American Advanced Study Institute on Transportation Sciences) award by the Rensselaer Polytechnic Institute. He is a registered Professional Engineer in the State of Maryland since 1997. Dr. Jha has served as a PI, Co-PI, or collaborator with other researchers on numerous research projects totaling over $4 million. The key sponsoring agencies of his research projects include Army Research Lab., Maryland State Highway Administration, Federal Highway Administration, National Science Foundation, and several Baltimore area consulting firms. Dr. Jha has authored (or co-authored) more than 80 articles in journals, books, and conference proceedings in the highway design, optimization, and transportation literature. He has also co-authored 2 text-books on road design entitled “Intelligent Road Design” and “Fundamentals of Road Design.” ISBN: 978-960-6766-87-9 9 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 ISBN: 978-960-6766-87-9 10 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 An urban designers’ road hierarchy TAEKE M. DE JONG Department of Architecture Delft University of Technology (TUD) Berlageweg 1 2628 GA Delft THE NETHERLANDS T.M.deJong@tudelft.nl http://team.bk.tudelft.nl Abstract: - This paper takes a factor 3 between the levels of a road hierarchy as a basis of a rough preliminary ideal typical network model for urban designers making proposals to be checked by traffic engineers. The spatial model starts to cover the area with potential connections, then leaving out levels or parts of the tissue in low density areas. It emphasises the total network rather than separate roads connecting poles of origins and destinations. The existing connections can be compared with the remaining connections in the model to interprete the actual levels. The model can be developed to predict congestion where levels are missing resulting in proposals for improvement or completion. Other interfering networks like waterways can be treated in the same way giving insight in the number of (expensive) crossings by the interference of different networks in different lay-outs. Key-Words: - road hierarchy, networks ,mesh size, urban design, capacity, intensity, congestion. All input figures in this paper are arbitrary. They illustrate a way of calculating rather than an always context sensitive reality. In the free downloadable original spreadsheet [4] they can be changed according to empirically verified context sensitive figures. So, the results can be checked for specific locations. Here the general inference is object of study and open to critique and improvements. 1 Introduction As a rule of thumb for urban designers I use a factor 3 between the levels of any road hierarchy since 1988 [1]. It was meant for educational purposes only. However, in 2000 traffic engineers concluded the proportion 3 is better than 2 or 4 for any of the three criteria they used as a starting point for calculation [2]. So, they questioned if the factor 3 might be a ‘natural phenomenon’. It raised the question if traffic jams in the Netherlands could be solved by adding regional highway levels between the local and national ones. This paper takes the factor 3 between the levels of a road hierarchy as a basis of a rough preliminary ideal typical network model for urban designers making proposals to be checked by traffic engineers. The spatial model covers the area fully with potential connections, leaving out levels or parts of the tissue in low density areas. It emphasises the total network rather than separate roads connecting poles of origins and destinations. The existing connections can be compared with the remaining connections in the model to interprete the actual levels. The model can be developed to predict congestion where levels are missing resulting in proposals for improvement or completion. Other interfering networks like waterways can be treated in the same way giving insight in the number of (expensive) crossings by the interference of different networks in different lay-outs [3]. 2 A factor 3 hierarchy If you leave your home to go for a ride, you start on a ‘residential street’ (for example 20m wide between the façades) via a larger ‘neighbourhood road’ (say 30m wide) reaching an even larger ‘district road’ (say 40m wide), and so on. At average (according to the starting point) every third road of each level you can make a turn to a road of a higher level (see Fig. 1). In urban design, the question arises at which mutual absolute distance you have to draw them. To keep it simple, we take a mesh width of 30m for the smallest residential paths, 100m for residential streets, 300m for neighbourhood roads, and 1000m for district roads (Fig. 1). That semi-logarithmic sequence I used earlier to distinguish orders of scale, based on the approximate radius of settlements of different order of hierarchy in the Netherlands through the ages [5]. In the Netherlands now it is often used to distinguish orders of scale avoiding scale falsification (the phenomenon that conclusions drawn on a specific level of scale can turn into their ISBN: 978-960-6766-87-9 11 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 opposite at another level of scale, already demonstrable at a linear factor 3) [6]. 3 Rectangularity higher level forced by a The most efficient enclosure surrounds the enclosed area with a minimum length of road (indicating investment). As well known, the result is a circle. In a continuous network it is approximated by a hexagonal pattern. This minimal ratio between periphery and area is demonstrated in many natural phenomena [7] (for example cells in a tissue) where preference is given to a minimal ratio between outer area and inner content. A good example is a cluster of soap bubbles. A cluster of soap bubbles forced into a thin layer (demonstrable by two overhead sheets) produces a two-dimensional variant. The bubbles arrange themselves in polygons with an average of six angles. However, if one pulls a thread through them, the nearest bubbles will re-arrange themselves into an orthogonal pattern (Fig. 2). Fig. 1 Five orders in a network hierarchy in nominal measures These measures are ‘nominal’, in a sense that ‘3’ may be interpreted between 1 and 10, ‘10’ between 3 and 30 and so on. The nominal (naming) character is expressed as (radius) R = {1,3,10 …}. Using a radius R means that the level of scale indicated by R also indicates the order of size of surfaces and volumes. So, R = {1,3,10km …} also indicates the orders of π·R2 surface of approximately {3, 30, 300km2 …}, at average increasing by a factor 10. If the capital R represents the largest radius (frame) taken into account, then r can be used as the smallest one (granule). The proportion r/R represents the resolution of a drawing. A proprotion 1/10 characterises a very rough sketch, 1/100 a sketch and 1/1000 a blue print. That proportion also indicates a tolerance by which a drawing or verbal argument may be interpreted. If the tolerance is supposed to be equal to the grain, Fig. 1 has a tolerance of 10m, because the smallest width of a drawn neigbourhood road (20m) has a radius of 10m. The model is rectangular and in Fig. 1 it is represented in the special case of squares. Firstly we will motivate rectangularity and then extend the model to elongated orthogonal variants with the same network density as the chosen squares. Hildebrandt and Tromba (1989) [8] Fig. 2 The formation of right angles Urban developments from radial to tangential can also be interpreted against this background. The interlocal connections ‘pull the radial system straight’. The additional demand for straight connections over a distance longer than between two side roads (a ‘stretch’) introduces rectangularity. Then every deflection from the orthogonal system is less efficient. This can be clarified without mathematics by a thought experiment. Imagine a rectangular framework with hinged corners that is completely filled with marbles. If one re-shapes this framework into an ever narrower parallelogram, then there will be space for fewer and fewer marbles. So, in every case, the rectangular shape proves to be optimal in this respect. A triangular grid has a larger ISBN: 978-960-6766-87-9 12 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 perifery/area ratio. If the parallelogram in the thought experiment that became ever more skew, matches an angle of 60° and an extra connecting line is added to cut the parallelogram into two equilateral triangles then it is an equilateral triangular grid with a much larger perifery/area ratio. width approaches its limit. In that case, while the net density is 2 km per km2, there can exist no ‘crossroads’ any more. That indicates elongating meshes saves the number of (expensive) crossings per km2! Class 1 2 3 4 5 6 7 8 residential path residential street neighbourhood road district road urban highway highway regional highway metropolitan highway directly served area m radius | mesh | crossing distance directly served number of inhabitants number of building layers along the road m width from façade to façade Fig. 4 Approximate spatial characteristics of a factor 3 road hierarchy 4 Fig. 3 Length (L) and width (W) of the mesh for a given net density of (D=2) Fig. 3 shows a sequence of relationships between mesh width and length in rectangular meshes with a net density of 2 km per km2 (the same density means the same investment). It is a common network density of country roads and district roads in the Netherlands, recognisable on any topographic map. Length and width of squares are both 2/density. However, the same density also occurs in a pattern of roads that go infinitely into one direction every 0.5 km. So, if the length and width of the mesh equals 1/d = 0.5 km, the ratio between length and A road hierarchy Going on that way we can make a table (Fig. 4) with approximate measures (in reality they will vary around that logarithmic average) for any type of road in a hierarchy. In Fig. 4 ‘m radius’ is a nominal measure for the area involved. It applies the mesh width of the theoretical network as well, the distance between crossings of roads of the same level (turn distance). ‘Directly served number of inhabitants’ is as elastic as the nominal surface (read 1000 inhabitants and think ‘something between 100 and 10 000 inhabitants’). Fig. 11 illustrates how much surface can be occupied by non residential functions. So, the density of in habitants as supposed in Fig. 4 should be adapted to the actual context. For example: a ISBN: 978-960-6766-87-9 13 ISSN 1790-2769 100 10 20 30 40 60 70 80 10 1 2 3 4 6 7 8 10 000 000 1000000 3000000 100000 10000 1000 100 10 100 000 10 000 30 000 3 000 1000 100 300 30 metropolitan region estate ensemble neighbourhood district conur-bation region town WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 region (R=30km) often has many open spaces, so there we counted 3 000 000 inhabitants instead of 10 000 000 as found in London and Paris. The spreadsheet mentioned in the introduction of this paper draws the adapted profiles as well according to the supposed number of dwelling layers, the required space for private gardens, pedestrians, parking, bicycle lanes, green verges and vehicle lanes filled in (Fig. 5). Traffic intensity The light vehicles/24 hour intensity is calculated in Fig. 7 by multiplying the number of directly served inhabitants, the number of car rides/inhabitant per day (multiplied by 2 for leaving and coming back) and the % infrastructure used surrounding the urban island of origin. For residential and neighbourhood roads there are visitors as well and the use of higher orders depends on the proportion of larger ride lengths. In the neighbourhood it does not count so much, but on roads of higher level, cargo transport and service traffic is more important because of their larger ride lenghts. For cargo we propose a simple, perhaps not yet reliable way. It is open for improvements, but it is used here to avoid complications, hiding the main line of inference. We estimated the kg cargo/ inhabitant/day and divided it by an estimated kg cargo/vehicle to get the number of cargo vehicles/24 hour. Class 1 2 3 4 5 6 7 8 directly served number of inhabitants 10000000 20000 0,00 20000 1000 1 200000 50% 0,02 1000000 0,10 50% 100000 1 1000 2000 0,00 2000 3000000 6000 0,00 6000 1000 1 150000 50% 0,05 100000 2000 0,01 200 1000 1 20000 50% 0,20 10000 400 0,02 20 1000 1 10000 50% 1,00 1000 20 0,01 2 1000 1 2000 50% 2,00 100 2,00 0,01 2 100 1 200 50% 2,00 10 0,20 0,01 2 10 1 20 50% 2,00 Fig. 5 Variants of 3 supposed profiles Traffic capacity The capacity is the maximum possible intensity. The ‘km/hour design velocity’ in Fig. 6 shows which speed of cars is supposed determining the ‘minimum lane width’ of the lanes as components of the roadway. The ‘number of lanes’ is determined by the expected number of cars per hour calculated in line ‘vehicles/hour intensity’ in Fig. 8. Class 1 2 3 4 5 6 7 8 110 130 km/hour design velocity 150 10 30 50 70 90 inhabitants car rides/inhabitant/day %surrounding infrastructure used light vehicles/24 hour intensity cargo kg cargo/inhabitant per day kg cargo/vehicle 1.75 2.25 2.75 3.25 3.25 3.25 3.25 minimum lane width number of lanes vehicle/hour capacity per lane 3.25 cargo vehicles/24 hour intensity 10 16 1 2 2 2 4 8 1000 1500 2000 2000 2000 2000 2000 500 service service visit/inhabitant/day 160 000 200 000 vehicle/24 hour capacityof the road 320 000 20 000 30 000 40 000 80 000 5 000 service vehicles/24 hour intensity Fig. 6 Approximate capacity of a factor 3 road hierarchy Fig. 7 Approximate intensities of a factor 3 road hierarchy ISBN: 978-960-6766-87-9 14 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 In a comparable way the number of service visits/inhabitant per day produces the service vehicles/24 hour intensity. Summing these lines, produces the number of vehicles/24 hour intensity (Fig. 8), which divided by 10 approximates the number of vehicles/hour intensity. If enterprises are taken as residents, the intermediate deliveries are taken into account as well, but in that case the number of rides per resident should be higher. However, the number of visits per ride could be more than 1 as is supposed here. The capacity is sometimes filled by the intensity in peak hours causing congestion. In Fig. 8 intensity and capacity are compared in the % use by car = intensity/capacity. Above a certain percentage (for example 60%) you can expect congestion. Class 1 2 3 4 5 6 7 8 104000 162000 vehicles/24 hour intensity 240000 10420 22200 2022 204 22 5 From a model into reality Now, we have to check how reliable this model is, knowing that reality always differs. We restrict ourselves to the spatial characteristics. A complete survey should take more cases into account and measure local intensities to check the theory. Here we take one case only and we do not check all assumptions. In Fig. 10, a map of the Dutch town Dordrecht, we find 6 levels of roads. The resolution does not permit to see residential paths (1). But in a coloured version we see residential streets (2, white) , neighbourhood roads (3, yellow), district roads (4, same colour, but somewhat thicker), urban highways (5, purple), highways (6, red), regional highways (7, red and orange). We have drawn circles of nominally 3, 1 and 0.3km in the map to indicate the size of parts we nowadays call city, district and neighbourhood. 10400 16200 81% vehicles/hour intensity % use by car = intensity/capacity 0,4% 1,0% 26% 28% 65% Fig. 8 Approximate intensity of a factor 3 road hierarchy Noise and house price The dB(A) noise on façades depends on many factors like intensity and distance to the façade. It is a rough estimate, but it determines the % devaluation of house prices by noise load [9]. However, living at a main road may have advantages. Fig. 10 The urban area around Dordrecht Class 1 2 3 4 5 6 7 8 75% 7% 24000 1042 2220 202 20 2 Deviation of predicted measures dB(A) noise on façade [10] % devaluation of houseprice by noise [9] Fig. 9 Approximate noise load and reduction of house price of a factor 3 road hierarchy Let us take Papendrecht (above right in Fig. 10). It has some clear squares of approximately 500x500m neighbourhood roads while our model states 300x300m. However, elsewhere (for example in the central part of the city), there are smaller mesh widths (sometimes 100m). So, the model fits better the average. Moreover, we appointed: “read 300m and say ‘something between 100m and 1000m’. So, reality deviates within the appointed tolerance of the model. If our model fits the average, we can say: “Papendrecht has a relatively large mesh width for its neighbourhood roads”. 66 59 62 74 80 84 90 48% 10% 5% 22% 34% 40% ISBN: 978-960-6766-87-9 54% 7% 96 15 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 Within the ‘neighbourhood’ square all streets are approximately 20m wide from façade to façade, according to what we supposed in Fig. 4. The neighbourhood roads fit the prediction to be some 30m wide as well. However, the district road is not 40m, but 50m wide. between the levels of a road hierarchy offers the opportunity to make a simple global traffic model stressing the network with all road categories rather than calculating the temporary requirements of separate roads. The model keeps us attentive on regularities in the existing urban tissue to be applied in urban design. It can globally predict traffic intensities and potentials of future connections without extensive research of origins and destinations between traffic poles. It is simple and visually convincing for administrators and designers, open to professional criticism, improvements and adaptations to local contexts. References: [1] Jong, T. M. de, Morfologische typologie van netwerken (Delft) Faculteit Bouwkunde TUD 1988 [2] Nes, R. v. and N. J. v. d. Zijpp, Scale-factor 3 for hierarchical road networks: a natural phenomenon? (Delft) Trail Research School Delft University of Technology, 2000 [3] Jong, T.M. de; Akker, C. van den; Bruin, D de; M.J., Moens.; Steenbergen, C.M.; Toorn, M.W.M. van den [eds.], Sun, wind, water, earth, life and living; legends for design. (Delft) TUD, Faculteit Bouwkunde, Publicatiebureau, 2007 [4] Downloadable from http://team.bk.tudelft.nl > Publications 2008 > traffic.xls [5] Jong, T.M. de, Milieudifferentiatie, een fundamenteel onderzoek (Den Haag, Delft) PhD Thesis Technische Hogeschool; Rijksplanologische Dienst, 1978 [6] Jong, T.M. de, Kleine methodologie voor ontwerpend onderzoek (Meppel) Boom; http://team.bk.tudelft.nl > Publications 1992 [7] d’Arcy Thomson, W., On growth and form (Cambridge UK) Cambridge University Press, 1961 [8]Stefan Hildebrandt and Anthony Tromba, Architectuur in de natuur, de weg naar de optimale vorm (Mathematics and optimal form), Wetenschappelijke Bibliotheek Natuur en Techniek, Maastricht/Brussel, 1989 [9]calculated according to Ruiter, E. P. J., The Great Canyon. Reclaiming land from urban traffic noise impact zones (Zoetermeer) Peutz b.v., 2004 [10]calculated according to the standarised Dutch model SRM1 http://www.stillerverkeer.nl/index.php?section =rmv&subject=RMW02&page=SRM1 Fig. 11 A Papendrecht detail There are two possible reasons of deviation: the spatial context and a superimposed function. Deviations caused by spatial context We supposed, there should be a district road every 1km, but in Papendrecht in Fig. 10 we see only one within a radius of 1km (diameter 2km). However, there is interference with the network of rivers, clarifying why the second one is not realized. A second one here would not have enough use to legitimate the cost. The river limits its supporting surface. The model supposes a homogeneous topography while reality is heterogeneous. Nevertheless the density of district roads is low comparing to the model, so the remaining one needs more capacity. Superposition A district road appearing in a grid of neighbourhood roads can take over a neighbourhood function (superposition). That is another reason to increase its capacity and thereby its width. 6 Conclusion A traffic network determines the lay-out of a city for a long time, covering changing requirements. Assuming a apparently timeless rational factor 3 ISBN: 978-960-6766-87-9 16 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 Urban design transformations reducing public paved surface TAEKE M. DE JONG Department of Architecture University of Technology Delft Berlageweg 1 2628 GA Delft THE NETHERLANDS T.M.deJong@tudelft.nl http://team.bk.tudelft.nl Abstract: - Public pavement for traffic and parking is expensive. It has to be paid by lots a land developer can sell, surrounded by public space making them accessible. This paper shows a systematic series of urban design transformations keeping as many variables constant as possible. With little difference in accessibility their urban effects are reported, in particular the impact on the surface needed for public paved surface. Key-Words: Urban design experiments, public paved surface, urban island, neighbourhood, district. the urban island; the North an South sides of an urban island are chosen because they are best suitable for parking for two reasons: their surface enlarges the NorthSouth distance between outer walls of dwellings, giving more acces to sunlight, and the shadow of North walls is welcome to parked cars; sidewalks are seldom smaller than 2m wide; there are no front gardens; dwellings are 5x10m, 2 floors high with roof timbers of 3m on lots of 100m2, housing 2.25 inhabitants in rows not exceeding 40m to avoid extra dilatation; paths around the back are 1m wide; supposing a location in the Northern hemisphere green areas are drawn East and South filling up the main 30m grid. They show the space saved by design operations (see Fig. 2), but they could be used to enlarge the lots for sale as well, diminishing public space (pavement + green) per lot. 1 Introduction Public pavement for traffic and parking is expensive. It has to be paid by lots a land developer can sell, surrounded by public space making them accessible. This paper shows a systematic series of urban design transformations keeping as many variables constant as possible. With little difference in accessibility their urban effects are reported, in particular the impact on the surface needed for public paved surface. 4. 5. 6. 2 The urban island The urban islands (the smallest urban ensembles completely surrounded by roads) drawn in this section are made by a free downloadble computer programme.1 The design transformations are mirroring, turning, repeating and elongating. Starting points for these experiments (to keep as many variables constant as possible) are: 1. centre lines of surrounding roads are chosen on a multiple of 30m (preliminary main grid); 2. roadways are everywhere 6m wide; that is not needed everywhere, but it includes a reservation for wider roads of higher level in the traffic network elsewhere; 3. the parking standard is everywhere more than 1 parking place per dwelling along the road, starting at least 5m from road corners, only drawn along roads North and South (indicated as ‘N’ and ‘S’) in the drawing of 7. 8. These starting points can be changed easily in the computer programme. However, for the time being they are kept constant below to study the change in allotment performance by the chosen design transformations. Mirroring the smallest urban island Fig. 1 shows an urban island with three houses with gardens surrounded by sidewalks and streets. North and South of the island there are parking lots for 2 cars each. The allotment is mirrored at the other side of the street. The ISBN: 978-960-6766-87-9 17 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 smallest urban island taken in consideration here has a grid measure of 30x30m. The consequence of small urban islands is an excessive surface of public pavement (here 76%!), leaving relatively little for sale (here maximally 22%) paying for that public space. Fig. 3 30x60m N-S Turning and repeating Fig. 4 30x60m N-S mirroring Fig. 1 30x30m Fig. 2 60x30m W-E mirroring The first design transformation, W-E mirroring in Fig. 2, elongates the urban island reducing public pavement (here into 67%). Now, W and E roads are used for entrance to houses at both sides and back gardens get more privacy. The lots for sale differentiate in morning~ and evening sun lovers. The gained surface is visualised by a green margin of 9m wide drawn East and 2.50m drawn South to get an idea of road profiles and crossings without greenery in the corner left below in the drawing, where circles are drawn with a radius of 10 and 30m.. Greenery is drawn East, but for children in the afternoon and in the summer evening green area can better be designed in the West to have sunny playgrounds. That does not change the land use. At a length of 60m, one side with parking places is enough to reach more than 1 parking place per dwelling. The shadow of the N side is best suitable for parking. N-S mirroring (Fig. 4) introduces North gardens, drawn longer here to get a partly sunny view on the N garden still. It differentiates the lots for sale in size and suggests a different dwelling type for sun lovers with south gardens and artistic life style with Northern light rooms like studios. Elongating To reach the same capacity of Fig. 4 by one sided elongating avoids the path round the back utilizing the side walk, giving back a proper size to the sidewalks N and S (Fig. 5). East gardens are suitable for people who like morning sun in the garden and in the sleeping room. Pavement is still 66%. Taking sun into account N-S turning and repeating (Fig. 3) gives both blocks South gardens. Now, the short sides of the urban island are used for parking, forcing cross-parking to reach >1 parking places per dwelling. The path round the back is enlarged at the expense of sidewalks to give proper front access to the Southern block. Fig. 5 30x60m elongating Fig. 6 60x60m mirroring Mirroring (Fig. 6) gives evening people a chance as well and both gardens more privacy. It differentiates use and plantation. ISBN: 978-960-6766-87-9 18 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 The enlargement of the urban island again reduces the amount of pavement into 52% in favour of the margins possibly used as green area: 9m East and 5m South. L-shape and U-shape Introducing perpendicular blocks (Fig. 7) provides streetcorners with front entrances in 2 directions. That gives the beginning of an urban look and safety by private control of public space on both roads involved. To improve that effect design solutions for corners, not implemented here, would be nice. Such solutions will struggle with smaller or no gardens in the corner. Fig. 9 90x60m Closing Fig. 10 Elongating and adapting 100x60m Violating the starting points, N-S elongating to 100m is easy by adding 2 houses West and East. However, the elongated blocks of Fig. 10 exceed 40m and need an extra dilatation, which is expensive. The shortage of parking places forces parking at all sides. By giving up cross parking N and S, there is space for 6 extra houses in total. The reduction of pavement is 2% only, but the number of parking places is 1.4 per dwelling. This time the green margin is distributed W and E to make trees possible. Fig. 7 30x60m Lshape Fig. 8 60x60m Ushape 500 450 400 350 m per lot 300 250 200 150 100 50 0 1 2 3 4 5 7 6 8 9 10 Fig. nr. of experiments (sorted) Mirroring the L-shape (Fig. 8) produces Ushaped allotments with one open side, here avoiding North gardens. It has the same advantages as previous mirroring transformations, in this case reducing pavement from 66% in Fig. 7 into 52% in (Fig. 8) and introduces green margins of 9m East and 5m South. S gardens go 0.5m around the back now, giving space for ivy-covered side façades avoiding grafitti. public green / lot public paved area / lot for sale / lot Closed urban islands Closing the urban island with front entrances on every surrounding street produces a usual allotment type of 90m length, leaving a 9m green margin East to fill the urban grid of multiples by 30m. Limiting parking places to N and S urges cross parking at both sides to have more than 1 parking place per dwelling leaving little space for sidewalks. Comparing land use Fig. 11 shows a sorted summary of the land use per lot for sale of the 10 reported ISBN: 978-960-6766-87-9 19 2 Fig. 11 Land use per lot of the experiments ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 experiments. In Fig. 10 we have left the starting points behind and start to look at a higher level. On that level larger spaces for mobility are needed (Fig. 12). Neighbourhood quarters A neighbourhood quarter of Fig. 12 does not only need extra space for pavement of neighbourhood roads, but also for neighbourhood facilities like green, water, perhaps a school, shops and offices. They produce car movements as well, but in the same time they are part of the modules involved unsuitable as residential area. Moreover, the area has to accommodate space for district roads (for example 40m wide, drawn left and below in Fig. 12). So, not all modules will reach 56 dwellings per ha or a floor space ratio (FSI) of 68% reached in Fig. 10. Suppose there are 500 inhabitants per neighbourhood quarter, it produces 500·2·2≅2000 car movements per day using half of the neighbourhood roads available. So, a neighbourhood road has some 1000 car movements per day or 100 per hour and much more in peak hours. 3 Traffic production If we repeat the module of Fig. 10 (100 x 60m) 5 times E-W and 3 times N-S (Fig. 12) we reach a mesh width (300m x 300m) suitable for neighbourhood roads. 4 Other transformations Fig. 12 A neighbourhood, repeating Fig. 10, reaching a radius of 300m (circles drawn with a radius of 100m and 300m) We now have 15 modules together surrounded by larger neighbourhood roads requiring extra space (for example 30m width between the facades). We call them ‘neighbourhood quarters’, because 4 of them make a neighbourhood (with a radius R=300m, for example supporting a primary school). Fig. 13 and Fig. 14 show two allotments of 100 dwellings ( 225 inhabitants) in rows of 10 on 1.8 ha. So, there are 56 dwellings/ha and FSI= 56% while the floor space per two storey dwelling is 100m2. From total area 63% surface is for sale and 37% is public space including 1 parking place per dwelling and roadway pavement of 4.2m wide. Residential roads Suppose every urban island contains some 75 people going out 4 times a day of which 3 by car. Suppose 1/3 of the car trips the driver is accompanied by a passenger, 1 trip is done by walking or cycling. So, the urban island produces 75·2·2≅300 car movements per day, because they are not only going out, they are coming back as well. That normally means 30 car movements per hour per island. Let them use only two of the four streets around the smallest island. So, a residential street has some 15 car movements per hour, but much more in peak hours. And there are visitors as well. Fig. 13 100 dwellings along residential streets with parking in front of the house: 32% pavement, 6% green area However, from the public space the proportion of pavement can be reduced in fabour of green. In Fig. 14 parking is concentrated at the boundaries. People have to walk 1 minute more than in Fig. 13 to reach their cars, partly living at residential paths, saving 45% of pavement (Fig. 15). However, vehicles should ISBN: 978-960-6766-87-9 20 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 have incidental access through the green area by reduced pavement. The disadvantage is, you can not easily come close to your home with luggage, moving vans and other vehicles. And you can not see your car from your home. Fig. 16 300m central road A central neighbourhood road costs least pavement, but it divides the neighbourhood and the school in two parts (Fig. 16). Fig. 14 Reduced pavement by residential paths, parking at 1 minute walk: saving 45% pavement, a profit of 350% green area That reduces municipal costs (or ground prices and taxes for private persons) substantially. By doing so, there is ample 3 times more green area (20% green of total area), resulting in a much greener look without cars. Part of that area could also be sold reducing municipal costs again. 20000 18000 16000 14000 12000 m2 10000 8000 6000 4000 2000 0 A B PPD-ZH(1970) Fig. 17 1800m peripheral one way road substituting 600m residential street A peripheral road costs much more road length, unless it is part of a grid used for adjacent neighbourhoods as well. A one way solution (Fig. 17) may half pavement and barrier effect but causes detours. green paved for sale Fig. 15 Reduction of street pavement, increase of green area comparing Fig. 13 and Fig. 14 Neighbourhoods Repeating a module like Fig. 14 by 8 around a centre, produces a neighbourhood of 1800 inhabitants, enough for some facilities like a primary school (1ha black square in Fig. 16 to Fig. 18), playgrounds, some shops and enterprises or public facilities. Fig. 18 900m peripheral road substituting 300m residential street, central parking ISBN: 978-960-6766-87-9 21 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 A one sided peripheral road (Fig. 18) leaves the other side open, but it causes long walking distances. Locating parking spaces at the boundaries of the ensembles, at daytime some residential parking space can be used by users of the facilities, avoiding extra facility parking space. Concentrated parking on neighbourhood level could mean a 10 minute walk to your car. However, these choices are often subordinate to the environment, mostly a district grid (Fig. 19). the same surface left (4·6=24) gives a more spacious effect located in the periphery (6·84·6=24)3. A positive effect is a better visual accessibility of the built-up area. Districts Repeating the module from Fig. 16 by 4 (7200 inhabitants) the surface fits in a 1·1km grid of district roads (40m wide), leaving open a 30m surrounding margin and a centre (Fig. 19) in each district quarter. That centre can be used for additional district green, facilities or housing (4ha black square), utilizing concentrated residential parking in day time. The grid permits to leave out 1200 m neighbourhood streets, but it will ask 8·90=720m extra residential roads to give access to all ensembles. By doing so X-crossings are replaced by Tcossings giving access roads along the square a view on larger buildings. Berlage designing the Mercator square in Amsterdam called it the ‘turbine principle’. The resulting T-crossings refer to the ideas of Camillo Sitte2. Fig. 20 Same built-up area optically full or empty 5 Conclusion Reducing the surface of public pavement by urban design could have positive side effects without loosing accessibility substantially. The first measure is to increase the surface of urban islands improving the area/circumference ratio. Secondly concentrating parking space could produce a substantial reducton of paved surface, applied on a lager scale partly by double use. Thirdly, parts of a grid could be left out replacing X-crossings by T-crossings. References 1 The computer programme is downloadable from http://team.bk.tudelft.nl > Publications 2003 2 Sitte, Camillo (1991) De stedebouw volgens zijn artistieke grondbeginselen (translation of the original 1889 publication Der Städtebau nach seine künstlerischen Grundsätzen (Rotterdam) Uitgeverij 010 3 Tummers, L. J. M. and J. M. TummersZuurmond (1997). Het land in de stad; de stedebouw van de grote agglomeratie (Bussum) THOTH. Fig. 19 A small district or district quarter Fig. 20 shows the optical principle of leaving the centre open, applied in Fig. 19 on the level of the quarters and on the level of its centre: ISBN: 978-960-6766-87-9 22 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 Developing a Mobile GIS for Field Geospatial Data Acquisition EBRAHIM POORAZIZI, ALI A. ALESHEIKH Department of GIS, Faculty of Geodesy and Geomatics Engineering, K.N. Toosi University of Technology ValiAsr Street., Mirdamad Cross, Tehran, P.C. 19967-15433 IRAN ebrahim.poorazizi@gmail.com, alesheikh@kntu.ac.ir Abstract: - Field based and GPS assisted GIS are frequently used in various geoscience applications. Such systems deliver more advanced, time and cost effective tools than traditional field forms for information collection. At the same time, different systems are on the market. These systems mostly enable users to determine positions of geo-objects by GPS, and maintain functionalities to gather and analyze geometric information. The digital obtaining of the application’s specific information is often underrepresented within these kinds of systems. This paper attempts to develop a system for gathering, integrating, analyzing, and visualizing spatial and aspatial information in electrical industry facilities. A PDA based GIS is extended to collect information in a beneficial manner, thus, adequately supporting the requirements of its users. Mobile GIS architectures, the components of a mobile GIS, and their impacts on gathering data are examined. The steps followed to design and develop a system for the electricity Transmission and Distribution Company are demonstrated. The results are presented and discussed in details in the paper. Key-Words: - Mobile GIS, GPS, PDA, Spatial information, Aspatial data 1 Introduction GIS databases, used by government agencies and private organizations, must be kept up-to-data to inform decision makers about proper management of resources [1, 2]. Data gathering is a time and money consuming task [3]. The recent evolutions in Mobile GIS caused the possibility to cost effectively gather, and manage GIS data [4]. Mobile GIS is the accessing, using, and storing of geospatial data directly in the field. Mobile GIS addresses the needs not only of GIS managers, but also of field inspectors, maintenance teams, utility crews, emergency repair workers, and other field workers who need timely access to GIS data in the field [5]. Mobile GIS is evolved with developments in:  Global Positioning System (GPS) technology  Rugged handheld computing technologies  Wireless communications, and  GIS software for mobile platforms Mobile GIS depends heavily on high quality data in the GIS database. However, in many organizations this data can be obsolete or even non-existent. GIS data collection is the process of populating a GIS with data on the properties, including position and attribute information, of interest to the organization. Until recently, collecting and using information in the field was a paper-based process with multiple points of data entry without accessing to real-time information or the ability to accurately communicate field observations back to the central stations. The recent developments in mobile GIS technologies have benefited many field-based information gathering. Information collection can now be performed more accurately with higher efficiency [6]. Mobile GIS can facilitate the following field processes:  Asset inventory—Recording the location and attribute information of an asset on a digital map  Asset maintenance—Managing asset location, condition, and maintenance schedules in the field  Inspections—Maintaining digital records of field assets for legal code compliance  Incident reporting—Spatially recording accidents or events. The process of decoding field notes, entering valid information into the geodatabase, and redistributing new paper maps is time-consuming, ISBN: 978-960-6766-87-9 23 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 expensive, and at risk for errors. However what is routine and necessary is to maintain geodatabases [7]. Mobile GIS has made it possible to digitally view and edit spatial data in the field. It overcomes the shortcomings of paper maps and provides real benefits to field-workers as well as cost benefits for the utility. A GPS data collection system has the same basic components as a Mobile GIS system. It requires the user to locate and then record position and attribute information about the features of interest. GPS technology provides the obvious choice for recording reliable position information, while handheld computer platforms running fieldoptimized software allow the user to efficiently record feature and attribute data [5]. During the past years, tremendous advances have taken place in GPS technology (receivers), data collection hardware, and field data collection software. Not only has the autonomous GPS accuracy improved, but the data collectors have become smaller, lighter, and less expensive. The software has become cheaper and easier to learn [8]. With the introduction of Palm Pilots followed by Microsoft's launch of a pocket PC operating system, a new generation of handheld Personal Digital Assistants (PDA's) have swamped the market. It is now possible to use these lightweight handheld PDA's, with GPS/GIS data collection software, for field applications. Onboard digital cameras allow users to include a visual record as part of the field data collection process. An application can automatically control the camera, preview an image, and finally take a proper photo for possible inspections. The photo can then be linked to its real world location where it was taken and associated with descriptive attribute information. In this paper a system is developed to assist operators in data collection for electrical power industries. of hand-held challenges. Various architectures have been proposed for mobile GIS implementation, namely; Stand-alone, Client-Server, Distributed, Services, and Peer-to-peer [11]. The simplest Mobile GIS architecture is the Stand-Alone Client. In this architecture the application and spatial data reside entirely on the mobile device (Figure 1). Although some applications may profit from this approach, such settings have major drawbacks. First, the hardware resources of the mobile device restrict the amount of spatial data the application can support. Second, this architecture does not allow for communication with any other applications. Fig.1: The Stand-Alone Client Architecture To address the above challenges, client-server architecture can be adopted. As Figure 2 presents, spatial data is moved to a separate computer and served to the client by GIS server software. The GIS applications can still be in the mobile device searching for data in data servers. Depending on the GIS applications, the location of the servers, and the network constraints, thin, thick or medium clients may be adopted [13]. In such architecture, the spatial data is constrained by the resources of an enterprise server. Moreover, multiple mobile devices running the same application can access the server concurrently, making this a potentially multi-user architecture. However, the architecture poses a new constraint: communication. If the mobile cannot establish communication to the data server, the GIS applications are useless, and the architecture losses its versatility. Due to range, and interferences, inconsistent communication occurs frequently in mobile applications. 2 Mobile GIS Architectures Mobile GIS is based on mobile computing and mobile Internet. It is an extension of Web GIS to mobile Internet including wireless Internet/Intranet and mobile communication network [9]. Mobile GIS have several restrictions due to the limited capabilities of both hand-held hardware and the network data transfer speeds and bandwidth [10]. The diversity of mobile devices, the limitation in processing power and screen display, and the diversity of mobile system platforms are examples Fig.2: The Client-Server Architecture In order to overcome the inconsistent connectivity, two challenges must be addressed: persistence and resource management. A distributed framework can handle the logic for persistence and resource management (Figure 3). ISBN: 978-960-6766-87-9 24 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 Fig.3: The Distributed Client-Server Architecture To further extend the back-end functionality of mobile GIS, applications can view the GIS server as a web service and allow for other web services to be part of the application (Figure 4). Since the web services use similar communications protocol, all mobile devices can communicate with each other. In addition, the web services can also communicate between themselves using SOAP XML, the industry standard for passing messages between software components [14]. Once employed, this architecture can support robust communication between any number of mobile devices and web services. Unfortunately, it might not be the best for some applications, such as those designed for collaboration in remote areas where connectivity to servers is unavailable. Fig.5: The Peer-to-Peer Architecture 3 Mobile GIS Components The core components of a mobile GIS are the same as generic mobile business systems. A mobile GIS has three fundamental components: hardware, software and the wireless network, which connects the mobile device to a data repository. The hardware component consists of the mobile device; a suitably configured wireless modem; a Web Server with wireless support, i.e., a WAP Gateway, a Communications Server and/or a Mobile Communications Server Switch so that the mobile device can communicate with the Internet or an Intranet; and an application or database server that contains the application’s logic and databases [16]. The software component includes the mobile device operating system (Windows 98/2000/NT, PalmOS, Win CE, EPOC, etc.); the mobile application user interface, which may run through an Internet browser; application server and/or database server software; application middleware if the mobile device needs to communicate with legacy (predecessor) systems or web-based application servers; and wireless middleware that links multiple types of wireless networks to application servers [17]. The wireless network component may be either a private network such as that used by law enforcement or emergency services, or a public shared network. Connectivity to wired networks or wireless LANs may also be included depending on the requirements of the mobile application. Fig.4: The Services Architecture In a peer-to-peer architecture a server is no longer available to keep spatial data; the data must be stored on the mobile devices themselves as seen in Figure 5. However, if each mobile device stores 100% of the data, then the architecture is restricted in the same manner as the Stand-Alone Client. To allow for more data storage throughout the application, each mobile device keeps just a subset of the data. When Mobile Device A needs data, it relies on its distributed framework’s resource management to know if it has that data locally. If it does not, the distributed framework must know how to access that data on Mobile Device B [15]. 4 Case study Since many organizations’ activities are based on spatial information, collecting spatial data has become outstandingly as one of the organizations target. Electricity Transmission and Distribution companies are among the companies that require spatial data and need a spatial information system with an up-to-date database of the conditions of the transmission lines, distribution lines, posts and other spatial features. By considering the importance and ISBN: 978-960-6766-87-9 25 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 the role of a spatial information system in the electrical industry (production, outfitting and maintenance parts), a system for collecting spatial data related to transmission and distribution posts was designed. A user-friendly system was developed to enable operators to gather and save descriptive information, the feature’s position by GPS, and the feature’s picture. The system is designed to work off-line (Stand-alone Client architecture) as well as online (Distributed architecture). In this project a HP iPAQ hw6945 was used to collect data. The device was equipped with an internal GPS receiver, and a 1.3 mega pixel camera. The device has also the features of a cell phone in addition to the ability to communicate via infrared, Bluetooth and Wi-Fi technology. The mobile is integrated with a high sensitivity GPS receiver and HP iPAQ Quick GPS Connection technology. Quick GPS Connection technology is a software application residing on the device that enables a faster connection for enhanced GPS performance. The operating system of the device is Windows Mobile and has the ability of programming (Application programming) using .net technology. The system’s software is consisted of two parts: the database and the user interface. The needed information contents for collecting geodata was extracted from the related data models and the database was designed in SQL Server CE (Compact Edition). The user interface was designed based on .net compact framework using Window Mobile APIs. A connection between the program and the database was established and the program was setup for use in Pocket PC. By using the designed software the ability to insert, edit and search information, to register the position and the picture of the feature is provided; in such a manner that the information will be saved in the database and its’ transmission and retrieval in the central database will be easily done. Figure 6 presents the several user interfaces for data entry and capturing photos. Fig.6: The user interfaces of the developed system The system has been set up for practical uses in Tehran Regional Power Company. The initial results are promising as it populates the database in real time. The deployed system has several practical features. The most compelling are:  access to data in the field (where it is often needed the most)  capture data in the field and in real-time (it includes photos)  append positional information to data capture  run GIS functionality in the field, (where again it is often needed the most) Further GIS applications are requested to guide the maintenance crew to the needed utilities locations. 5 Conclusion Both the hardware and software available for digital data acquisition have advanced considerably in recent years. Professional clients look towards mobile platforms to increase productivity through efficient information handling, resulting in cost reduction, as well as a well-informed mobile workforce. This paper analyzed mobile GIS architectures and its components. The paper also presented a mobile GIS for power industry users. The ability of the system to capture positional data through its GPS receiver, gather descriptive information using its API, and take photo made the system cost effective. With mobile GIS, data is always in a digital format, making it easy and efficient to transfer from the field to the office without introducing interpretive errors. Checks and balances are still required, of course, but many of these can be automated so those carrying out the checks can focus on the real errors. ISBN: 978-960-6766-87-9 26 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 References: [1] Alesheikh, Ali A., and F. Sadeghi Naeeni Fard, Design and Implementation of a Knowledge Based System to Improve Maximum Likelihood Classification Accuracy, Canadian Journal of Remote Sensing, Vol. 33, No. 6, 2007, pp. 459467 [2] Longley, P. A., M. F. Goodchild, D. J. Maguire and D. W. Rhind, Geographic Information Systems and Sciences, Wiley Publications, 2005 [3] Hosseinali, Farhad and Ali A. Alesheikh, Weighting Spatial Information in GIS for Copper Mining Exploration, American Journal of Applied Sciences. Vol. 5, No.9, 2008, pp. 11871198 [4] Amirian, Pouria, and Ali A. Alesheikh, A Hybrid Architecture for Implementing Efficient Geospatial Web Services: Integrating .NET Remoting and Web Services Technologies, Journal of Applied Sciences, Vol. 8, No.5, 2008, pp. 730-742 [5] McLarin, Pat, Mobile GIS Technology Update, Proceedings of Map Asia 2004. [6] ESRI, ArcPad: Mobile GIS, ESRI White Paper, September 2004 [7] Larry Wilke, Redline Map Editing Goes Digital, Energy Currents, GIS for Energy, ESRI, fall 2003 [8] Wadhwani, A., Recent advances in mobile GPS/GIS mapping technology, GISdevelopment.net, http://www.gisdevelopment.net/technology/mobi lemapping/techmp005pf.htm (last viewed at March 12, 2008) [9] Fangxiong, W., and J. Zhiyong, Research on a Distributed Architecture of Mobile GIS based on WAP, Proceedings of ISPRS, Commission VI, Vol. XXXV, July 2004 [10] Jane Drummond, Roland Billen, Elsa João, and David Forrest, Dynamic and Mobile GIS: Investigating Changes in Space and Time, CRC Press, Taylor & Francis Group, 2006 [11] Bryan G. Hassin, Mobile GIS: How to get there from here, R7 Solutions, Inc, 2001 [12] Robert Laberg, Srdjan Vujosevic, Building PDA Database for Wireless and Mobile Development, Wiley Publishing, Inc., 2003 [13] Alesheikh, Ali A., and H. Helali, Web GIS Development Strategy, GIM International, Vol. 16, No 11, 2002, Pp 12-15 [14] Alesheikh, Ali A., A. K. Oskouei, F. Atabi, and H. Helali, Providing Interoperability for Air Quality in-Situ Sensors Observations Using GML Technology, International Journal of Environmental Science and Technology, Vol. 2, No. 2, 2005, Pp 133-140 [15] Kalantari, S.M., Ali A Alesheikh, and A. Talebzadeh, Promoting Distributed GIServices Using Mobile Agent Technology, Proceedings of MapAsia 2003 Conference, Kuala Lumpur, Malaysia [16] Amirian, Pouria, and Ali A. Alesheikh, Publishing Geospatial Data through Geospatial Web Service and XML Database System, Journal of Applied Sciences, 2008, (In press) [17] MobileInfo.com, Mobile Computing Hardware & Software Components, http://www.mobileinfo.com/hwsw_components. htm (last viewed at Mach 14, 2008) ISBN: 978-960-6766-87-9 27 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 Developing a Genetic Algorithm for Solving Shortest Path Problem SAEED BEHZADI, and ALI A. ALESHEIKH Department of GIS, Faculty of Geodesy and Geomatics Engineering, K.N. Toosi University of Technology Valiasr Street, Mirdamad Cross, Tehran, P.C. 19967-15433 IRAN alesheikh@kntu.ac.ir Email: behzadi@sina.kntu.ac.ir , Abstract: Routing is gaining prime importance in our increasingly mobile and highly Information Technology enabled world. The shortest path problem is widely applied in transportation, communication and computer networks. It addresses the challenges of determining a path with minimum distance, time or cost from a source to the destination. This paper introduces a novel Genetic Algorithm (GA) approach to solve the shortest path problem. A connected graph is assumed in which all the weights of the paths are positive. New mutations as well as crossover parameters are defined. The use of mutation parameters depends on the extent of the area under consideration, and the distance between the start and end points. In order to evaluate the proposed algorithm, an urban road map is selected. In all case studies, the algorithm was successful in determining the shortest path. The details of the experimental results are discussed and presented in the paper. Key-Words: Genetic Algorithm, Shortest path problem, Mutation, Crossover, Graph theory. 1 Introduction The problem of estimating a shortest path between two nodes is a well-known problem in network analysis. Shortest path algorithms are subject of extensive research, resulting in a number of approaches for various conditions and constraints [2, 4, 5]. The shortest path problem; finding the path with minimum distance, time or cost from a source to a destination; is one of the most fundamental problem in transportation networks. It arises in a wide variety of scientific and engineering problem settings, both as stand-alone models and as subproblems in more complex problem settings [1]. This paper presents a new Genetic Algorithm to solve the shortest path problem. Practical examples can be found in finding the shortest path (i.e. the shortest possible distance) between two nodes in a graph or two cities in a map with potential connections (assuming that the path distances are always positive). Typically, a transportation network is represented by a graph with each node representing a city and each edge being a path between two cities. Genetic algorithm is appealing as a solution, since it deviates from traditional algorithms that try to compare every possibility to find the best solution that might be a time consuming algorithm for a graph containing a large number of nodes and edges. Genetic algorithms are inspired by Darwin's theory about evolution. The genetic algorithm is an optimization solution that is based on natural selection. The genetic algorithm repeatedly changes a population of individual solutions [13]. At each step, the genetic algorithm chooses individuals randomly from the current population to be parents and uses them to reproduce the children for the next generation. Over successive generations, the population "evolves" to an optimal solution [7, 11]. The algorithm is started with a set of solutions (represented by chromosomes) called population. Solutions from one population are taken and used to form a new population. This is motivated by a wish, that the new population will be better than the old one. Solutions which are selected to form new solutions (offspring) are selected according to their fitness - the more suitable they are the more chances they have to reproduce. Basically, several random sets of parameters are considered for an algorithm, and a fitness value (optimization value) is calculated for each. Based on the fitness values, the best sets are mixed (Selection, Crossover and Mutation are combined) together and new sets are again applied to the algorithm until an optimal parameter(s) is obtained. This effect is usually obtained by breaking the genetic algorithm into a few small parts [10]. The algorithm stops when predefined conditions (for example the number of populations or improvement of the best solution) are met. 2 Genetic Algorithm ISBN: 978-960-6766-87-9 28 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 3 Implementation Each node in the map is given a unique integer value index from 1 to i, where, i is the number of nodes in the map. Each individual is designed to represent a solution for the problem and it does not have repeated node indices. The length of the individual is selected to be equal to the number of nodes in the total map, because there may be cases such that the shortest path may contain the total number. For a map with i number of nodes the gene length is equal to i, where Tj is the jth node in the map, namely; Route_1: [T1, T2… Tj,…T (i-1), Ti] Route_1 is a path that contains all nodes in the graph. Chromosome_Length: The length of each Chromosome in the population N_P: The number of points in the graph D (Start, End): The distance between Start and End points Area: the area of the graph that is approximately computed by using minimum and maximum of x,y in the area. This component is defined as (2): Area = ( Max _ X − Min _ X ) × ( Max _ Y − Min _ Y ) ( 2) 3.1 Coding Permutation encoding is used in this problem. In permutation encoding, every chromosome is a string of numbers, which represents the number of nodes in a sequence. The number of nodes in each chromosome is not equal. Integer vector P = (v1, v2, . . ., vk) is a chromosome to represent a path from nodes 1 and n. Because different paths include different nodes and arcs, the dimension of chromosome is not fixed. If (v1, v2, . . ., vk) represents a path from nodes 1 to n, then the path includes (1,v1)∈A, (v2,v3) ∈A, …, (vk,n) ∈A [12] then Xi,j is defined as:. ⎧1, ⎪1, ⎪ xi , j = ⎨ ⎪1, ⎪0, ⎩ if i = 1, j = v1 , if there exists l such that i = v1 , j = v l+1 , if i = v k , j = n, otherwise, n: The flexibility of the length of the chromosome. It is approximately defined as an integer number between 1 and 4 that depends on the density of the points between start and end points. It is 4 when the density of the points between start and end points is more than other parts. The first and last genes of each individual are consecutively the start and end points. Other genes in each chromosome are chosen randomly from the other points in the graph as they are not equal. 3.3 Fitness Function The fitness function in this research is defined as [1]: F ( x) = 1 − The _ number _ disconnect ed _ path Actual _ length _ of _ the _ path (3) For all (i,j)∈ A. It is also easy to verify that {xij ⎜(i,j) ∈A} obtained by this way is a path from nodes 1 to n. Conversely, let {xij ⎜(i,j) ∈A} be a path from nodes 1 to n. A chromosome is obtained by the following procedure. Considering the first part of the function, Actual_length_of_the_path is calculated by summing the route distances of the path. Therefore, the value for Actual_length_of_the_path is between 0 and 1. The_number_disconnected_paths counts the number of unreal path in the total path. Hence, the value for The_number_disconnected_paths is between 0 and i. It means that the function varies from -i to +1. If there is not a path between node t1 and t2 then the SLD (Straight Line Distance) between t1 and t2 is considered as the actual distance. 3.4 Fitness Scaling 3.2 Initial population The size of the population depends on the number of nodes in the graph and the length of each chromosome in the population. It is defined as: Chromosome_Length = N_P × D (Start, End) Area ×n (1) Where: Fitness scaling converts the raw fitness scores that are returned by the fitness function to values in a range that is suitable for the selection function. The selection function uses the scaled fitness values to select the parents of the next generation. The selection function assigns a higher probability of selection to individuals with higher scaled values [10]. After creating the initial population, fitness values for each individual are calculated. In this research, ISBN: 978-960-6766-87-9 29 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 proportional scaling is used to make the scaled value of an individual proportional to its raw fitness score. 3.5 Genetic Operator Genetic operators mimic the process of heredity of genes to generate new offspring at each generation and play a very significant role in genetic algorithm [8]. In our algorithm, the crossover operator, mutation operator and selection are defined as: 3.5.1 Crossover operator Crossover examines the current solutions to find better ones [6, 9]. Physically, crossover in the routing problems plays the role of exchanging each partial route of the two chosen chromosomes in such a manner that the offspring produced by the crossover will only be one route. This enforces the selection of one-point crossover as a good candidate scheme for the proposed GA. One partial route connects the source node to an intermediate node and the other partial route joins the intermediate node to the destination node [3]. ′ ′ ′ P = (v1 , v2 ,...vk ′ ) ; be two 1 Let P = (v1 , v2 ,...vk ) and 2 chromosomes. The crossover operation is: if there are common nodes between them, then randomly ′ choose one, say vi = vi′ . The following two chromosomes are then produced: ′ ′ ′ (v1 , v2 ,..., vi , vi′′ +1 ,..., vk ′ ), (v1 , v2 ,..., vi′′ , vi +1 ,..., vk ) Which are also feasible chromosomes representing paths from nodes 1 to n. If there is no common node, then nothing is done [12]. This process is shown in Fig. 1. Fig. 1. Crossover operator 3.5.2 Mutation operator In this research, three kinds of mutation are considered. The logic of using each kind of these mutation operators depends on the extension of the area and the distance between start and destination points. These kinds of mutation are defined as: 3.5.2.1 Internal mutation In this case, mutation operator is performed by randomly choosing two chromosomes in the new generation to mutate. Then, the two selected genes will be exchanged. For example (1 2 3 4 5 6 8 9 7) is a path that contains 1, 2,…, 8, 9, 7 nodes. By using this operator two nodes are chosen (2 and 8) then they are changed. (1 2 3 4 5 6 8 9 7) => (1 8 3 4 5 6 2 9 7) 3.5.2.2 External mutation In this case, mutation operator is accomplished by randomly choosing one chromosome in the new generation to mutate. The selected gene will then be exchanged with another gene that was not in the path. For example (1 2 3 4 5 6) is a path that contains 1, 2,…, 6 nodes. By using this operator one node is chosen (2) then it is changed with an external point (9) that was not in the path. (1 2 3 4 5 6) => (1 9 3 4 5 6) 3.5.2.3 Omitting mutation In this case, mutation operator is done by randomly choosing one chromosome in the new generation to mutate. Then, the selected gene is omitted. For ISBN: 978-960-6766-87-9 30 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 example (1 2 3 4 5 6 8 9 7) is a path that contains 1, 2,…, 8, 9, 7 nodes. By using this operator one node is chosen (2) then it is omitted. (1 2 3 4 5 6 8 9 7) => (1 2 3 4 5 6 8 9 7) => (1 2 3 4 6 8 9 7) The use of each kind of the mutation operators depends on the problem at hand. For example if the distance between start and end point is short, i.e. be more specific omitted mutation is preferred and if the distance between start and end point is long provide a number, the ration of using internal and external mutation is more than omitting ones. These kinds of mutations are shown in Fig. 2. Fig. 2. a. Internal operation b. External operation c. Omitted operation considered as weights of the arcs. The objective was to minimize the total weight of path. The result of this experiment is shown in Fig. 3. 4 Experiments To evaluate the performance of the outlined method, we performed experiments using actual road maps of parts of Tehran, the capital city of Iran (Figure.3). In this problem, the lengths of the arcs are (a) (b) Figure3. a. A part of the tested map b. Start and End point and the shortest path crossover and mutation operator are defined for the problem but in this paper one crossover and three kinds of mutation operator were defined to solve the 4 Conclusions and recommendations The main advantage of genetic algorithms is their flexibility. In most GAs, it is common that only one ISBN: 978-960-6766-87-9 31 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 shortest path problem. These operators are executed on a network. In each problem the crossover operator is always used but the ration of using each kind of these mutation operators depends on the extension of the area and the distance between start and destination points. By using this method the solution efficiency becomes higher. Other crossover operations can be used in addition to the ones introduced in here. Such crossover operations and the use of each kind of the operators can be defined base on the specific problem like what is done for mutation operation. For increasing the efficiency of this algorithm, the first population can be defined base on the direction between start and end point. In this case, not only are the nodes of each gene chosen randomly but also they are restricted between start and end points. design—an empirical study of the factors that influence performance, IEEE Trans. Evol.Comput. Vol. , No. 3, 2001, pp. 236–249. [9] X. Hue,. Genetic Algorithms for Optimization: Bachground and Applications. Edinburgh Parallel Computing Centre, Univ. Edinburgh, Scotland, Ver 1.0. , 1997 [10] http://www.mathworks.com/access/helpdesk/hel p/toolbox/gads/ [11] D. Whitly, A Genetic Algorithm Tutoral, Computer Science Department, Colorada State University Fort Collins, CO 80523 [12] Ji. Xiaoyu , Models and algorithm for stochastic shortest path problem, Applied Mathematics and Computation Vol. 170, 2005, pp. 503–514 [13] F. Hosseinali, and Ali A. Alesheikh, . Weighting Spatial Information in GIS for Copper Mining Exploration, American Journal of Applied Sciences. Vol. 5, No. 9, 2008, pp. 11871198. References: [1] S. Abeysundara, B. Giritharan, S. Kodithuwakku, A Genetic Algorithm Approach to Solve the Shortest Path Problem for Road Maps, Proceedings of the International Conference on Information and Automation, 2005, pp. 272-275. [2] C. Davies, P. Lingras, Genetic algorithms for rerouting shortest paths in dynamic and stochastic networks, European Journal of Operational Research Vol. 144, 2003, pp. 27–38 [3] M. R. Delavar, F. Samadzadegan, P. Pahlavani, A GIS-Assisted Optimal Urban Route Finding Approach Based On Genetic Algorithms, Proceedings of the ISPRS, 2004, Istanbul, Turkey. [4] E.W. Dijkstra, A note on two papers in connection with graphs, Numeriske Mathematics Vol. 1, 1959, pp. 269–271. [5] R.W. Floyd R.W., Algorithm 97: Shortest paths, Communications of the ACM, Vol. 5, 1962, 345. [6] D.E. Goldberg, Genetic Algorithms in Search, Optimization and Machine Learning. Addison Welsey Publishing Company , 1989. [7] J. Holland, Adaptatin in Natural and Artificial System, University of Michigan Press, Ann Arbor, MI, 1975. [8] C. Hsinghua, G. Premkumar, C. Chu, Genetic algorithms for communications network ISBN: 978-960-6766-87-9 32 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 NEW ECONOMIC DISTRICTS IN BELGRADE AREA Slavka Zeković, PhD, Tamara Maričić, MSc Institute for Architecture and Town& Regional Planning of Serbia, 11000 Belgrade, Bulevar kralja Alexandra 73/II SERBIA e-mail:zeksbmv@eunet.yu; www.iaus.org.yu Abstract: Dynamic changes in spatial structure of cities are in the focus of attention, as well as, the development of new economic zones/poles in urban areas, new spatial forms of economic activity. It is considered that the processes of globalization and market mechanisms have led to the ’break down’ of urban structures into numerous specialized localities, by way of developing economic clusters and other activities. The development of new economic poles in cities is a consequence of activating new localities and changing the present territorial organization and the development of ’knowledge-based’ economic activities. In city Belgrade the new economic districts – new industrial, commercial, entrepreneurial zones developed systematically or spontaneously in the suburban areas (along highways) have a major role in spatial development of economic activities and in the planning of territorial organization of that metropolitan areas. Key words are: new economic districts, spatial clusters, business activities, sustainable development, urban policy, the urban development zones/clusters within urban areas in Serbia. Therefore, this work is trying to demonstrate the need for the research of new economic forms in urban areas, the need for mechanisms of agglomerating activities in spatial/economic clusters and the harmonisation of sustainable urban development in Serbia, based on the example of the Belgrade area. 1 Introduction For an effective planning of sustainable development and spatial organization in urban areas in Serbia in the following period, it is necessary to incorporate European strategic frameworks, approaches and planning practises [1,2,3]. In the process of economic and social transition in Serbia, coordination to the conditions of EU competitiveness, the development of small and medium-sized entreprises in the industrial sector, services and other business activities, as well as urban policy, is a complex economic and urban planning challenge. Under the pressure of global processes in economic development, the transition of the socio-economic system into a market-oriented economy in Serbia, among other things, has influenced the creation of new economic poles/zones/clusters in urban areas, changes in spatial organization of cities, the apearance of new locational forms of industries, services, business activities, etc. The document Serbia’s Strategy of Industrial Development 2007-2012 [4], does not deal with the questions about the creation of new economic 2 Sustainable Urban Development The latest discussions concerning potential spatial development are not possible without taking into consideration the visions of socioeconomic development. However, in economic theory the ideas of certain schools about the vision of economic development have always been divided between free market and/or state regulation. Two key global tendencies have influenced the socio-economic and spatial changes – the globalization of economy and the transformation of postcommunist economic systems and state into a market economy, political pluralism and the strenghtening of institutional frameworks. The concept of ISBN: 978-960-6766-87-9 33 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 sustainable development, as a challenge in harmonizing economic, social, political, environmental and spatial dimensions could serve as a suitable frame for “depreciation of influences“ of globalization processes and socio-economic transitions on all levels of planning. Due to the influences of the latter processes, spatial organization of cities and settlements, regardless of big regional differences, is characterised by a ’Planetary“ sindrom of standardizing lifestyles and organization of work for people, together with a homogenization of urban structures and processes (the so-called European monotopy [5], in the sense of unification of places, spatial structures, the expanding new economic zones in cities areas). According to Jakšić M. [6], the challenge of the 21st century is not in establishng a fixed and final utopia, but in creating an ev-topia. In other words, instead of u-topia – the creation of an ev-topia (a place that evolves, develops), meaning a system of stimulating and applying knowledge and adjusting skills to the conditions, uncertainties and aims of the surroundings. As the mechanisms of perception and acquiring knowledge have a social character, social relations are based on territory, so is the development of economy based on knowledge togeter with mechanisms of urban planning policies. Principles of sustainable development should be added to these (CEMAT, 2000) [7]. 2 The Identification of Expanding Economic Activities in Urban Areas In the Green Paper on EU cities [8], the basic aims of urban development have been defined, based on the improvement of the quality of environment in these territories: (1) environment protection and management, which means reduction of uncontrolled pressure and growth of various activities, with the restoration of cities; (2) the curtailment of city participation in the causes of pollution, which means careful planning of the economic activity expansion and the use of spaces in the city that have been already ruined by devastated industrial and commercial objects. It is often the case that such objects in the cities are abandoned because of old technology and for other reasons, so it has been suggested that these spaces (brownfields) are put to good use by reconstructing them. The experiences of urban planning in Eastern–European cities have been directed more towards new green-fields and less towards activating brown-fields. Apart from this general European trend of curtailing regional differences, establishing new „economic poles of development“ in metropolitan peripheries is also significant in planning spatial development. According to Burdach J.[9], it is a matter of a new discourse in peripheral growth (metropolitan). New economic poles in metropolitan areas are a result of a high participation of the public sector ( especially in providing heavy infrastructure, support in curtailing spatial unbalance, etc), but also in attracting foreign and local investments. In some large cities of Europe, and Serbia as well, (ex. Belgrade, Novi Sad), new economic poles – new economic, commercial, industrial, entrepreneurial zones that have been created by planning or spontaneously in the suburbia (along motorways, main roads) have a priority in the development and spatial organization planning of the area. The reasons for such a trend are manifold – low price of land, available sites, proximity of residential areas, favourable conditions on site etc. The expansion of work/factory zones in big cities is contradictory to the idea of a sustainable compact city, above all, because of an increase in transportation, greater energy consumption, greater costs of infrastructure, negative effects on the environment, ruin of agricultural land and similar. In this way, the tendency of deurbanization has transformed into suburbanization, because the density of population in peripheral metropoliten areas has rapidly grown, as well as the number of flats, the growth of economic activity, costs of infrastructure, ecology etc. New centres of production and consumption influence the transformation of suburbia ( as mainly residential, socially homogenous zones, with lower density in an urban periphery) into post- ISBN: 978-960-6766-87-9 34 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 suburbia ( which expresses the transformational process in multi-functional locations). Many different concepts have been concerned with this phenomenon of the transformation of suburbia into post-suburbia, describing it as a „new centrality“ outside of the central place, i.e. the creation of a new centre outside the downtown city area. The term „new economic pole“ implies various kinds of new dynamic centres with a functional specialization in the metropolitan periphery. The main spatial forms of new economic poles in peripheral urban areas (suburbia) are industrial parks, technological parks, industrial complexes, shopping malls, business-commercial centres, logistics centres, business centres etc. According to [9], the concept of classic spatial models of cities (standardized „rings“ and sectors) is being more and more transformed into polycentric forms, created by grouping or networking different kinds of locations for different purposes. A tendency of the breaking up of urban structures into different series of specialized and fragmented localities, by way of clusters of activities disperesed inside a populated structure. In that way, more and more an image of a „functional archipelago“ is created in an urban (periphery) fabric, unlike earlier approaches. (For exampke, in earlier Master plan of Belgrade in 1970’s, the concept of an „archipelago in a sea of green“ was promoted). The cummulative effects of developing new poles lead to a new concept of growth of urban/metropolitan periphery as well. Initial nucleuses of this development are often shopping centres, business-commercial centres et sl., which is a consequence of the transition into post-industrial society. A significant part of urban spaces is occupied by industrial and transportational functions and facilities, often they are very negative. Recently, with privatisation, the process of reactivating abandoned industrial locations (brownfields) in eastern-european cities, including Belgrade is gradually being opened. In cities with a market economy, the industry holds 4-10% (Paris 5.2%, London 4,7%) of the entire developed space [10], while industrial locations in eastern-european cities occupy 15,1-43,8% (Prag 13,4%, Varšava 15,1%, Sofija 27,1%, Ljubljana 27,4%, Moscow 31,6%, St. Petersburg 43,8%. In Belgrade , it is approximately 18%. However, the general opinion is that due to the price of construction, parking problems et sl., the new dominant trend is construction in new industrial facilities, on free locations in the urban periphery (greenfields). Clustery deconcentration of business activity leads to a new relation and movement from the centre of city to the outskirts. The effects of development and concentration of industrial activity and living in the suburbia (post-suburbia), without control of the overconstruction phenomenon or urban lots, environmental effects and pollution et sl., have an ever-growing, partly explosive character. Post-socialistic „boom“ of the metropolitan periphery is not imminent to only easterneuropean and Balkan countries, but to developed metropolitan areas of Western Europe too. The development is shifting from the central compact city nucleus to the inner and outer city peripheries. One of the principle reasons for the socio-economic and spatial transformation of urban areas is the proces of tertialization, i.e. the development of services. The main instigator of these processes are usually foreign investments into the services sector (most often it is the banking sector, insurance, shopping malls, sales and exhibit halls, hotels, storehouses and trade etc). The role of city authorities and local investors in this proces is relatively small, usually they have a service function in providing suitable conditions on locations. The formation of new economic poles as a result of a general tendency for an international shift of production and services from the city centre to the periphery. In other words, the market mechanisms and factors of international proportions, activate the pressure of direct foreign investments into metropolitan/urban peripheries, above all, because of the agglomeration economies, reduction of various costs, acceptable and favourable locational economies in periphery city zones etc. This process has negative repercussions both in spatial-environmental, and in the institutional domain as well. Based ISBN: 978-960-6766-87-9 35 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 on theoretical opinions, experiences from many areas, it seems that the process has a devastating effect on the regional and local institutions, as well as on the local investors, by imposing on them the rules of behaviour, standards, movements and direction of capital. Inside an urban-spatial context, it can be directly observed in the profound changes ( quite often in the caving in as well) of the existing spatial organization of a city, city zonings, propositions, rules and regulation standards for using the building land et sl. So, direct foreign investment is the pivot of the development of new economic poles in urban areas (banking, shopping malls, high-tech and business activities, industrial parks, logistics centres and transportation etc). This process has a foothold in the theoretical concept of liberal economy, especially the so-called Smith’s „invisible hands“ of the market. In other words, the processes of illegal constuction and expansion of cities are only a consequence of bad legal solutions in the field of planning and building of spaces, poverty of citizens, social and other problems, but they are directly fuelled and/or initiated by market mechanisms. The process of suburbanization is „artificially“ initiated by economic and social policies as well, but also by inadequate mesures of urban policy and policy of urban land ( ex. untransformed system of managing the building land, undeveloped instruments of taxing building land and real estate, tax rates, the fee for land development and usage, local taxes, subventions, concessions, etc.) In accordance with economic restructuring (tendency towards tertialization) in urban/metropolitan structure, a stagnation and „disappearance“ of classic industrial zones, complexes, entreprises is evident. A functional conversion of these zones is evident, fuelled on one side by the process of privatization of state entreprises in these zones, and on the other side, by the pressure of direct foreign investment. According to [11], the post-communist deveopment of eastern-European cities shows a hybrid layout with relics of spatial structures of the socialist era, a phenomenon of structure transformation and new suburban/posturban spatial layout of clusters. They are nucleuses of new employment growth in city peripheries, and the first early signal of a polycentric structure of a territory. Spatial economy of periphery urban area is not homogenous. Although, until recently, traditional city peripheries were identified as a mixture of industrial spaces, family homes, traffic corridors and greenery, today they have a more distinct sensitivity to market signals and initiatives in relation to the central city zones. In this space, drawn by the growth of population with a higher education, especially by way of a new infrastructure for research-development institutions, many new high-tech activities of production take place services. There activities of transportation services are developed, logistics, production, wholesale (warehouses, storehouses etc.) shopping malls and services. 3 Results and Discuss of New Trends of Economic Activities in Belgrade Metropolitan The process of post-suburbanization is a consequence of activity of commercial powers and it takes place in all metropolitan peripheries, including Belgrade. Significant foreign investments and the development of 86,000 entreprises ( of which around 1/3 are in Belgrade) and 90,000 enterpreneurs, illustrate a more significant role of market mechanisms of allocated new economic content in the metropolitan area. According to Serbia’s Strategy for Joining the EU [12], a development of industrial parks is predicted, which would later grow into clusters, with the provision of necessary heavy infrastructure, possibility of fast construction of business and industrial facilities, fiscal incentives and qualified labour. The National Investment Plan of RS predicts the construction of 49 industrial zones in towns of Serbia. According to data [13], only in the area of Belgrade, in a „new wave“ of construction 20,000 ha of urban land (farmland) has been found to be under construction in the peripheral area. A housing deficit in Belgrade, numerous refugees and dislocated persons, have caused a significant residential pressure onto the suburbs, and ISBN: 978-960-6766-87-9 36 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 uncultivated farmland of Belgrade agglomeration (ex. Zemun corridor, belt of motorways to Surčin, Batajnica, Novi Sad, Avala direction, Borča, Ovča, Pančevo direction etc.) as well as the birth of new small entreprises, and the concentration of economic activity in the mentioned spaces. Metropolitan periphery outside the borders of the city of Belgrade is becoming more attractive for settlements (ex. the zones along the motorway Belgrade-Novi Sad, Belgrade-Zagreb, Belgrade-Niš, Ibar direction, Avala direction, Zrenjanin direction etc.) due to easy access to the corridors, nature etc. In addition, a concentration of economic activity is evident along the motorway from Belgrade to Batajnica, Novi Sad, Novi Beograd, airport “Nikola Tesla”, Dobanovci, Zemun, Pančevo road etc. A great concentration of economic activity has occured outside the Belgrade agglomeration , on the motorway zones - ex. large industrial zones in Šimanovci (550 ha), Pećinci (500 ha), Krnješevci (250 ha), etc. At motorway exits (corridor X) big shopping centres have been built like „Metro“, „Tempo“, „Idea“, „Rodić“, „Mercur“, „Mercator“, „Veropoulos“, etc. Municipalities that have better traffic and communication links with their surroundings, and with the central zone of Belgrade, and have an efficient entrepreneurial local authority and administration, are advantaged in attracting new content. In the Belgrade agglomeration, those are municipalities Pećinci (1,050 ha, Surčin (250 ha), Stara Pazova (1,900 ha), Indjija (930 ha) and others. At the Belgrade metropolitan area are located 12 industrial zones (about 2,500 ha). The process of globalization and the activity of the „invisible hand“ on the market, among other things, is the product of profound spatial, structural, urban and socio-economic changes on all levels. On the territory of Serbia, due to the activity of transitional changes these processes have been significantly boosted. However, they flow spontaneously, randomly and often without adequate planning and management or institutional control. In the period between 1990-2000, due to the many political economic and social happenings in Serbia, „grey“ economy was especially promoted, with an explosion of illegal construction and an expansion of cities. The aim of urban planning policy and the concept of clustery deconcentration activities is to prevent the negative consequences of the doom scenario and entropy of urban territories (conversion of the „boom“ development scenario into the so-called „doom scenario“), based on the priciples of sustainable development. Considering the fact that the development of new economic polarities in urban peripheries mainly is not linked to regional and national politics, the concept of polycentric urban structure could alleviate the negative effects of the mentioned tendencies. The development of potential implications of new polarities onto the regional environmnet and development, the manner of coexistence of growth and stagnation areas and/or depression, the disappearance of traditional industrial production, expansion of services, explosive growth of suburbia, is the subject of planning policy of metropoliten area. The current developing tendencies in the Belgrade metropolitan reflect the growing socio-economic differences through a special „functional archipelago“. In accordance with the experiences of the European metropolitan peripheries, the possible scenarios for the future expansion of new economic polarities in the metropolitan area of Belgrade: (1)an actual existence of area growth, mainly along important traffic corridors nearby the city centre, will continue in the future as well. Today this process is visible along the motorways Belgrade-NoviSad, BelgradeZagreb, next to the airports „Nikola Tesla“,etc; (2) the development of new industrial structures (modern production complexes – industrial parks, technology parks entrepreneurial zones, complexes) mainly by government aid, the arrival of multinational companies, the support of EU et sl. Apart from technological complexes, the new polarities are the shopping malls, distribution-transportation centres, business centres and others, (3) the prediction that the residential zone of suburbanization will expand in the future onto the green zones and will demonstrate the adverse side of suburbanization; (4) the process of ISBN: 978-960-6766-87-9 37 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 reurbanization in the Belgrade agglomeration is present sporadically, by way of rehabilitation of the existing industrial zones; (5) the development of new industrial/business spatial forms of the new economic poles of development in the urban and rural area, with the possible development of a big theme park. The establishment of airport city bussiness zone in New Belgrade, construction of Waterland in New Belgrade, construction of Aqualand nearby Dobanovci, along the motorway Belgrade-Zagreb, zones of recreation and entretainment like specific economic poles, (6)and apart from the problems and impossibilities of an accurate prognosis for directing the dynamics and developing processes, it is predicted that the centre of trends in the following decade in the Belgrade metropolitan will move to te periphery. Conclusions The development of economic activities within the new economic zones/poles/clusters in urban areas can be assessed as a consequence of the process of globalization. That zones could be some kind of mixure of old spatial structures. and the development of new locational models in the urban fabric, under the influence of market and the globalization process of economic activity and investment. In this process are evident: the existence of growth area, mainly along important traffic corridors nearby the heart of the city; development of new locational forms (industrial park, technology park, high-tech corridors), mainly with government aid, foreign investment, etc; large changes onto urban structures of metropolitan areas; over-construction of building locations; the effect on the price fluctuation of real estate in certain parts of metropolitan land; population mobility from the centre to the periphery. It is estimated that the absence of harmonization of European development policies, urban development, environment in the economic development in cities of Serbia could have consequences in the growth of competitiveness of urban economy and sustainable urban development. Regional Planning on 7-8 September 2000, Hannover, Strasbourg, Council of Europe [8]Green Paper on the Urban Environment, European Commission, 1990 [9]Burdach J.,New economic poles in European metropolitan peripheries: Introductory remarks on theory and empirical evidence, European Spatial reseach and policy, Vol.13 No 2/2006. [10] Bertaud A., The Spatial Structures of Central and Eastern European cities, in The Urban Mosaic of Post-Socialist Europe, Space, Institutions and Policy, Verlag, Springer, 2006. [11] Dovenyi Z., Kovacs Z., Budapest-The Post-socialist Metropolitan Periphery between ‘Catching up’ and Individual Development Path, ESRP, vol.13, No2/2006. [12] Strategy of the Serbia’s entry into the EU, (2005), Government of the Republic Serbia, Belgrade [13] Serbia Investment Climate Assessment, Finance and Private Sector Development Unit (ECSPF), (2004), Europe and Central Asia, World Bank References [1]Lisbon -revisited - Finding a New Path to European Growth’, Working Paper, 2004., EPC [2]Competitiveness and Innovation framework Programme 2007-2013, (2006), European Parliament and the Council, 1639/2006/EC, 24.October 2006, OJ L 310/15,09.11.2006 [3]Communication ‘Fostering Structural Change: an Industrial policy for an Enlarged Europe’, COM (2004), 274, 20.april 2004. EC [4] Serbia’s Strategy of Industrial Development 2007-2012, 2006, Government of the Serbia, Belgrade [5] Jensen O.B., Richardson T. (2004) ,Making European Space, Routledge, London [6] Jaksic M., Vision, institions and economic development, Ekonomski anali No.163/2004, Economics faculty, Belgrade [7]CEMAT ,Guiding principles for sustainable spatial development of the European continent, Adopted at the 12th Session of the European Conference of Ministers responsible for ISBN: 978-960-6766-87-9 38 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 SUSTAINABLE TRANSPORT PLANNING: A METHODOLOGICAL APPROACH Jorge Díaz Tejada Universidad Autónoma del Estado de Morelos, México. Abstract: Sustainability is a goal that no one yet knows how to achieve. The act of sustainable planning and design is a heuristic process; that is, one in which we learn by doing, observing, and recording the changing conditions and consequences of our actions. Observation, recording, and monitoring are crucial elements of the sustainable design process. Therefore, our data comes mainly from the on-site analysis, which is supported on some theoretical basis, as following mentioned. From the sustainable perspective, we may recognize the following premises as basic ones. The key to sustainable design is the system approach-sometimes called a holistic view. The second premise on which sustainable design is base is that product and process are one. Therefore, the process by which an end is achieved is often given as much, or nearly as much, weight as the product, because it is recognized that only by changing the design process is it possible to change the design result1 . • INTRODUCTION From the very beginning, we should consider to shift our more conventional planning approach, based on: Single Disciplinary theoretical basis (Quantitative basis) • • Obsolete Paradigms (Utilitarian function) Vertical Management (Social exclusive) As stated before, observation, recording, and monitoring are crucial elements of the sustainable design process, in this sense I consider essential to incorporate data from the observatories that have been 1 implemented in most cities lately , so that we can recognize main trends in terms of land use growth (residential, retail, etc), its impact on travel patterns, values related to environmental sustainability, safety, health, community sense, household income spent on transportation, etc. We can not assume a city as a neutral recipient, of homogeneous nature and consistency, it is instead a more complex phenomenon, multi1 Cross-Sectorial Management (Social inclusive) To a more inclusive model, based on: • Multi-Disciplinary theoretical basis (Quantitative, Qualitative and Relational basis) New Paradigms (Valued-Based) • The United Nations Settlement Program (UN-HABITAT) established in 1997 the initiative to raise urban observatories worldwide for monitoring some sustainability-related data. ISBN: 978-960-6766-87-9 39 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 dimensional (economic, social, environmental) which reacts to changing conditions, takes advantages of synergies, and where social behaviors (of non-linear nature) are a key driven force of mobility and transportation trends. The multi-dimensional character of the planning process can be recognised from the definition of Sustainable Transportation itself 2 A sustainable system is one that: • transportation PROJECT IMPLEMENTATION In order to link theoretical approach to the practice in terms of project implementation, I propose the frame shown in figure 2, where three main elements are suggested and are explained as follows: Theoretical Basis While software development, technical tools, modelling and cutting edge technology, in general have promote a sense of confidentialness, when applied for transport planning, I will remark some of the weakness and even misuse that this approach could lead us to. We have to think about the paradigm which is rooted in this approach. (Epistemological Nature) ¿How do we conceive the city? ¿Can modernism 3 still explains how the city functions? ¿Is the primarily function of the city a utilitarian one? • • Allows the basic access needs of individuals and societies to be met safely and in a manner consistent with human and ecosystem health, and with equity within and between generations. Is affordable, operates efficiently, offers choice of transport mode, and supports a vibrant economy. Limits emissions and waste within the planet’s ability to absorb them, minimizes consumption of non-renewable resources, limits consumption of renewable resources to the sustainable yield level, reuses and recycles its components, and minimizes the use of land and the production of noise. As a synthesis of what has been stated, we can look at figure 1, which in turn is useful to show the methodological frame as a reference for further explanation to be developed. 3 2 The Centre for Sustainable Transportation, Ontario, Canada. Report October 2002. As a movement in the arts, 1929, from modern (q.v.). The word dates to 1737 in the sense of "deviation from the ancient and classical manner" [Johnson, who calls it "a word invented by Swift"]. It has been used in theology since 1901. ISBN: 978-960-6766-87-9 40 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 Holistic Approach Process Output Sustainable Transport Economy Society Environment Government Non Government Society Figure 1: Methodological Frame Since Le Corbusier pointed out in the Athens’ Charta in 1933, that a city could be conceived as having three main functions (Residential, Work and Recreational) which are connected by the transport system, dramatic changes have occurred, not only in terms of explosive urban sprawl, extremely densified urban areas, megalopolis formation, but also in terms of social disequilibrium, environmental risk, safety, all of which have had an enormous impact (externalities) on transport cost, because of new ISBN: 978-960-6766-87-9 41 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 technologies needed, travel time, health deterioration because of pollution, sedentary style of life derived from car dependence, among many other consequences. On the other hand, we have to think about the strategies and operative tools derived from this paradigm. (Methodological Nature) Quantitative tools have been the dominant basis in research, as much as, for plan implementation up to now, while qualitative explanation has remained underestimated for researchers, authorities, and the community in general. I will remark some of the limitations when applying these tools alone, and which are related to the following issues: a. Linear behaviour logic b. Model calibration c. Variables Included d. Static forecasting e. No alternative scenarios The logic, which explains how a city functions, is a phenomenon of a very complex nature. We have to think about some reversible processes that have arouse as the target for transport planning, in most countries, which do not respond to a linear logic, or even some new strategies with no precedents, as following mentioned. • • • • • Lessen Car Dependence Pedestrianization Mixed-Land Use Traffic Tariffing Traffic Scheduling causing forecasting to have a wider range of uncertainty. Cross-disciplinary approach is not just an option but it is indispensable if we look for a neither neutral, nor idealized scenario, but a more comprehensible, social behaviour related territory. Some of the most noticeable issues that arouse when thinking of the city as a more complex phenomenon are: 1. Social Behaviour is a key-driven force for changing conditions. 2. Synergies between Transport and Land Use, have some generalized features but there are others which are related to local conditions. 3. Transport should focus in moving people not in moving vehicles, so pedestrian circuits besides nonmotorized modes should be included when planning the transport system. 4. Different territorial scales (Regional, Urban, Local) demand different transportation solutions (Edges, Nodes, Corridors) and modal integration (Rail, Subway, Buses) 5. Public Space is not just for transportation purposes but for social integration, leisure, diverse cultural manifestation. 6. Landscape should be considered as part of the transportation system. Society In the last few years, there has been a shift in the way different society groups are being incorporated during the planning process of the transportation system. Urban Observatories are not collecting data in a mechanical way, just to add some more information to justify a Besides this, mathematical models still recognize some variables that are not measurable or are of a non-linear nature ISBN: 978-960-6766-87-9 42 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 decision, but instead they are trying to perceive some value-based variables that different society groups consider extremely important and that are related to the concept of life quality. We shall distinguish different society groups, as following suggested: 1. Users of the Transportation system 2. Transport Operators 3. Non-Government Institutions 4. Government Authorities We still need to narrow our definition of these groups, as far as, quite different needs and interest are promoted within these groups, as following suggested. 1. Users of the Transportation system • • • Gender : Male/Female Age Groups : Children/Adults/Elder Activity : School Students/Workers/Housewives/Retir ed People We can figure out, how difficult could be to deal with different interpretations of the problem in a phenomenon of extremely complexity as it is the city itself, where disequilibrium, irregular processes, different income groups, or even different cultural beliefs are expressed. Empirical evidence4 has shown that dialogue between all these different social groups is essential if a plan for an improved transportation system is to be implemented. There is also an indispensable consideration to be made, the one related to Viability of achieving the plan in terms of the budget available, its cost and benefits related analysis, the time needed for building all the facilities, the urban fabric, geographical features of the site, and so on. To awake public awareness about the consequences of not doing anything, is of course a good way to begin promoting the change. Modal alternatives, opportunity of choice, connectivity between modes, nonmotorized facilities, pedestrianization, public space availability, are among others attractive ways of revitalize and improve not only the transportation system performance but above all a good way of getting a better life quality. 2. Transport Operators • • • • Private Companies Public Companies Local Services Regional Service 3. Non-Government Institutions • • • Local Consultants International Consultants Academy 4. Government Authorities • • • Local Authorities Federal Authorities Central Government Authorities Mexico City and Santiago de Chile City, exemplify the different outcomes derived from a successful process in the former, while an unsuccessful negotiation occurred in the later. 4 ISBN: 978-960-6766-87-9 43 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 Holistic Approach Theoretical Basis Society Territory Science Humanities Governmt. Non-gov. Regional. Urban Society Local Pilot Project Figure 2: Project Implementation. Territory Phenomena as; Megalopolis formation, conurbation, globalized metropolis, thematic city, dormitory cities, or land use patterns as; Shopping Malls, tourist corridors, industrial parks, make us thing about quite different transportation needs to attend to. ISBN: 978-960-6766-87-9 44 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 Nevertheless, we should consider all of them as part of a comprehensible transportation system. We can, of course, recognize at first three different territorial and functional scales: • • • Regional (Megalopolis, Conurbation) Urban (Metropolis, City, Town) Local (Historic city center, a particular street, business district, etc.) Nodes are extensive pedestrian zones which primarily function is to serve as transfer stations, with intensive public space use, leisure amenities, among other possible features. Considering as a whole, inter-modal connectivity is essential if we look for an improved transportation service. Despite the fact of considering it, an edge, corridor or node, an important issue to plan is to have alternative modal choices and not just one, so that vulnerability of a collapse will be of a lower probability. We should incorporate two concepts besides what has been described before. • • Complete Neighborhood Complete Street In correspondence with each one of them, I suggest the following concepts as part of the transportation system facilities. • • • Edges Corridors Nodes Edges are to connect different cities and are usually supply with a train system facility besides the toll highway system (and a toll-free highway if available). In the case of Mexico City, travels between the capital city and surrounding cities as Toluca (69.8Km), Cuernavaca (85Km), have been steadily growing in the last few years. A remarkable issue from this is that these trips are made on a daily basis, typically from home to work, indicating a strong functional link between these cities. Corridors are to connect different parts of the city, typically linking zones with the highest daily ridership. Bus Rapid Transit is a good example of a transport facility implemented in the last two decades in developing countries, as much as, in developed ones. In the case of Mexico City, it has been implemented the Metrobus (A BRT facility), along the Insurgentes Avenue (the longest avenue of the city). A complete neighborhood is a mixedland use area of high density, including different kinds of dwellings, health care services, schools, shopping center, leisure amenities and workplaces, so that the majority of people living there can also work in the neighborhood. The idea is to allow, as much as possible, walking or travel on non-motorized modes within the neighborhood, minimizing transit ridership. A complete street is a road where all (or most of) the transport modes are simultaneously included, so that people are able to choose the most suitable mode for their personal purposes. These modes may include in different lanes: • • • • • A Subway A BRT A Cycling way A Walkway A lane for cars ISBN: 978-960-6766-87-9 45 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 REFERENCES Thompson F. George and Steiner R. Frederick. Ecological Design and Planning. John Wiley &Sons, INC.USA.2002 The Centre for Sustainable Transportation, Ontario, Canada. Report October 2002. Amar, Georges. Movilidad Urbana Sustentable: Un Cambio de Paradigma.2007 Willumsen, Luis. Regulación y Control del Transporte Colectivo: Lecciones de Transantiago en Chile.2007 Schulze, Oliver. Reorganización del espacio, transporte no motorizado y accesibilidad.2007 Kogan, Jorge. Tendencias y Retos de la movilidad en America Latina.2007 Replogle, Michael. Lograr Transporte de Alto Rendimiento.2007 AUTHOR BIOGRAPHIES Diaz, Jorge. Graduated as a Bachelor in Architecture in 1988, before being awarded a Master’s Degree in Urban Planning in 1991 and a PhD in Urban Studies in 2004. Membership: World Scientific and Engineering Academy and Society (WSEAS)-USA. International Institute for Mobility Training (IIFM)-Switzerland. International Council for Canadian Studies (ICCS)-Canada. Sistema Nacional de Evaluación Científica y Tecnológica (SINECYTCONACYT)-México. He has more than fifteen years of experience as a consultant for the government of México. He has been teaching and developing research for the last ten years. He has recently been awarded a Postdoctoral research fellowship by the embassy of Canada. Copyright License Agreement The Author allows ARRB Group Ltd to publish the work/s submitted for the 23rd ARRB Conference, granting ARRB the non-exclusive right to: • publish the work in printed format • publish the work in electronic format • publish the work online. The author retains the right to use their work, illustrations (line art, photographs, figures, plates) and research data in their own future works The Author warrants that they are entitled to deal with the Intellectual Property Rights in the works submitted, including clearing all third party intellectual property rights and obtaining formal permission from their respective institutions or employers before submission, where necessary. ISBN: 978-960-6766-87-9 46 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 Environmental Quality of Roads in Heraklion, Crete ANDRONIKI TSOUHLARAKI Lecturer Technical University of Crete Kounoupidiana, Hania, GR-73100, Greece Department of Environmental Engineering email: niki.tsouchlaraki@enveng.tuc.gr ELEFTHERIA ZOAKI Civil Engineer, Aristotle University of Thessaloniki MSc, Environmental Design of Cities and Buildings ΜSc, Environmental Design of Infrastructure Projects 7, Kipselis rd., Heraklion, GR-71409, Greece email: zoritsa70@yahoo.com Abstract: This paper analyzes the parameters that determine and affect the environmental quality of urban roads. These parameters are grouped in five categories: bio-climatic, urban planning, traffic parameters, artificial elements and other parameters. After theoretically analysing all the parameters of the above categories, they are tested in practice for 16 representative roads of the urban complex of Heraklion. Following this, we compare the results of the above-mentioned analysis that can be quantified and draw certain conclusions for the current environmental quality of the roads examined and, by extension, for all the roads of the urban complex of Heraklion. Based on these conclusions, we propose certain interventions for upgrading the environmental quality of the city's roads that are of relevance to the state, as well as to the private sector. Keywords: environmental quality, urban roads, bioclimatic, urban planning, traffic parameters, Heraklion 1 Introduction The public, open areas of a city, i.e. the roads, squares, gardens, churchyards, archaeological sites, are the ones that make up the picture of a city, with the figures, the dimensions, the analogies and the perspectives that they create and the buildings that surround them. It is exactly the picture experienced and perceived by the resident/citizen and the picture gathered by the visitor, who returns to his or her hometown [1]. The majority of citizens are used to consider the public space as being a remainder of the private space, as a resulting space and not as a space designed. "People draw blocks in the form of regular shapes and they name the remainder roads and squares", writes C. Siite. In this way, the road continues to be an area of movement, and not an area of transport, in which the notion of the interim space disappears, transferring the importance towards the starting and the ending points of the path [2]. The par excellence social area of the city, the road, is substantial only when it offers the minimum environmental conditions that will allow residents to move safely, as well as enjoyably [3]. Today, the urban road is the infrastructure for vehicles (moving and stationing), people (as an area for playing, meeting, demonstrating, speeching, walking, parading, socialising, etc.), as well as for cyclists and public transport. Depending on the values of the individual society, its economic and cultural level, as well as on the technology available, the same area can be distributed differently among the various uses and users. In addition to the road, per se, its equipment is also an important means which, when used ISBN: 978-960-6766-87-9 47 properly, contributes positively to the formation of the road environment, while being also directly related to the mood and the picture finally gathered by the user (pedestrian, driver, passenger, permanent resident, visitor, etc.). Another important parameter is also the materials used for roads and pavements, as well as for roadside buildings, as they affect the microclimate, the amenities of open areas, and the energy efficiency of the surrounding buildings. 2 Factors - indices of environmental quality of urban roads As mentioned above, the factors-indices of the environmental quality of the urban roads examined are grouped into five categories, as illustrated in the table below: Sunshine - Sun protection Ventilation - Wind protection Bio-climatic Air pollution Parameters Noise Visual disturbance Building permit limits building heights Building density and system Urban Planning Road orientation Parameters profiles Building forms Border spaces Pavements ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 - with a direct impact: Pedestrian streets One-/two-way roads Cycle lanes - with an indirect impact: Traffic load Traffic Parameters Traffic makeup Average movement speed Parking locations and patterns Longitudinal slopes Road pavement materials, road surface condition Roadside construction materials Artificial Elements Road equipment Artificial lighting Green elements Population density Inward and outward movements Means of transport User safety Cleanliness and hygiene Other Parameters Preservation, rehabilitation and reuse of historical and cultural resources Land value Table 1: Environmental quality indices Examining the environmental parameters for all of the roads of Heraklion's urban complex would be a time-consuming effort, without any particular aim. For this reason, we selected roads representative of different areas of the urban complex in Heraklion, i.e. roads with different urban planning parameters, orientation and, therefore, bioclimatic parameters, as well as with different traffic patterns and technical characteristics, together with roads presenting different mixtures of such parameters, so that the description of the current situation, and the proposals for improvement that will be developed later, reflect as far as possible, the entirety of the cases presented in the city. The map below depicts the urban complex of Heraklion and the roads selected. Fig. 1: Locations of roads examined The following findings are derived from the analysis of data for each road: 2.1 Sunshine – Sun protection 21 ΔΕΚΕΜΒΡΙΟΥ - ΗΛΙΑΣΜΟΣ 9% 29% ΚΑΘΟΛΟΥ ΜΕΡΟΣ ΠΛΗΡΗΣ 62% Fig. 2: Sunshine Sixty-two percent (62%) of the roads examined are fully shaded during the winter months, either because of their orientation, or because of the height and location of buildings on both sides of the road, the width of the road or a combination of all these factors. 21 ΙΟΥΝΙΟΥ - ΗΛΙΟΠΡΟΣΤΑΣΙΑ 11% 1% ΚΑΘΟΛΟΥ ΜΕΡΟΣ ΠΛΗΡΗΣ 88% Fig. 3: Sun protection Eighty-eight percent (88%) of the roads examined are fully sunlit during the summer months, either because of their orientation, or because of the height and location of buildings on ISBN: 978-960-6766-87-9 48 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 both sides of the road, the width of the road or a combination of all these factors. It is, therefore, an urgent need to shade these roads during the summer months, utilising means and techniques, as presented below. The buildings on both sides of the roads examined are two-storey (29.4%) and single-storey constructions (25.1%). In other words, more than half of the buildings (54.5%) have a height below 6.00 meters. 2.2 Ventilation– Wind protection ΑΕΡΙΣΜΟΣ 2.4 Height/width ratio ΛΟΓΟΣ υ/π 18% 38% ΕΠΑΡΚΗΣ ΑΝΕΠΑΡΚΗΣ 62% 27% <0,4 0,4-0,7 >0,7 55% Fig. 4: Ventilation Thirty-two percent (32%) of the total length of the roads examined are ventilated sufficiently, owning to their orientation, which matches (perfectly or closely) the direction of the prevailing winds. Seventy percent (70%) of the said length can present phenomena of channelling, with eddies and swirls and, for this reason, it is considered necessary to find solutions in order to address this problem. Most of the roads in the urban complex of Heraklion have a N-NE orientation, given that, as a historical city, Heraklion follows the radial system of roads, i.e. a pattern that favours the ventilation of large road sections, during most days of the year. This orientation coincides with the direction of the winds prevailing in the area. The remaining road sections favour the entrapment of pollutants, a problem which can nevertheless be prevented with the use of solutions and techniques presented below. Fig. 6: h/w ratio Despite the fact that 82% of the total length of the roads examined has a h/w ratio <0.7, as shown above, the road sections in question do not present ideal bioclimatic conditions. This can be explained by the fact that the literature on the conditions of environmental quality in relation to the h/w ratio focuses more on the buildings on either sides of the road, than on the road itself. The existence of border spaces in areas built using either the "open on all sides" or "open" construction pattern increases the distances between the buildings and provides better conditions of hygiene and safety and visual and thermal convenience, for the users of both the road and the roadside facilities. 2.4.1 In relation to the building permit limits ΛΟΓΟΣ υ/π ΣΕ ΣΧΕΣΗ ΜΕ ΤΟ Σ.Δ. 2.3 Building heights 2,8 1,8 <0,4 1,2 0,4-0,7 >0,7 7,05% 15,58% 25,12% ισόγεια 2ώροφα 3ώροφα 4ώροφα 0,8 0,6 22,86% 29,39% >4ώροφα 0% 20% 40% 60% 80% 100% Fig. 7: h/w ratio, in relation to the building permit limits Fig. 5: Building heights As one would expect, areas with a high building permit limit (bpl factor of 2.8 and 1.8, respectively) exceed the set h/w ratio of 0.7, as happens also in areas of new expansions, in which ISBN: 978-960-6766-87-9 49 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 small properties and illegal constructions prevail, regardless of the building permit limit which has not been respected. It is known for these areas, to "have been built first, and thereafter planned in urban terms". High bpl values (2.8 for the central area and 1.8 for its periphery) have been adopted to a large extent. It is also commonplace in some plots, the constructions built before the enactment of these bpls (i.e. before 1978) to exceed the bpl and, in some cases, remarkably. 2.4.2 In relation to the building system ΛΟΓΟΣ υ/π ΣΕ ΣΧΕΣΗ ΜΕ ΤΟ ΣΥΣΤΗΜΑ ΔΟΜΗΣΗΣ 2.5 Longitudinal slopes ΚΑΤΑ ΜΗΚΟΣ ΚΛΙΣΕΙΣ 26% 0-2% 2-4% >4% 57% 17% Fig. 9: Longitudinal slopes In the roads examined, and by extension in the urban complex of Heraklion, the relief is smooth, with slopes ranging between 0 and 2 percent for 50% of the roads and a mere 26 percent of the roads presenting intense slopes (>4%). Therefore, pedestrian movements are favoured, as is also cycling. Furthermore, the production of air pollutants and noise from vehicles is not increased. Σ <0,4 Π.Ε. 0,4-0,7 >0,7 Ε 2.6 Means of transport 20% 40% 60% 80% 100% 0% Fig. 8: h/w ratio in relation to the building system As pertains to the building system, values above 0.7 for the h/w ratio, are remarked in the continuous building system (narrow roads, especially in the historical part of the city and/or tall buildings) and in the areas of illegal constructions (over-exploitation of the private spaces to the detriment of public space). The continuous building system, typical of the central area and its greater periphery, results into the entrapment of air pollutants and noise. The scaled arrangement of the last storeys of buildings, in areas with such a pattern, partially mitigates the above problems and positively affects the sunshine on the road. Both the "open-on-all-sides" construction pattern of the old city expansions, and the "open" construction pattern of the new expansions, together with the relatively big intervals between the buildings allow for better sunshine conditions, as well as for visual and thermal convenience, compared to the central area. Owning a private car was once an indication of prosperity and rise of the living standards. Nowadays, the widespread use of the private car is beyond any control, given that the purchase of a car is affordable for all. Thus, all of the roads examined and, by extension, in the entire city, the majority of journeys are made by private car, including short distance journeys that could otherwise be made on foot. A parameter contributing to the excessive use of the private car is the problematic public transport network. In many cases, the organisation of bus routes and itineraries takes place on a rather empirical basis, resulting in frequent deviations from the scheduled itineraries. Even in areas covered sufficiently by the public transport network, there is lack of sheltered bus stops, that are safe for awaiting passengers. Also, many areas, especially in the new city expansions, are not included in the bus route network. Finally, the compulsory use of the city centre by all the bus routes is an obstruct to, and is obstructed by, the general traffic conditions. Obvious is also the lack of a peripheral route connecting areas outside the city centre, something that otulines the lack of a suitable peripheral road axis connecting the out-of-centre areas. 2.7 Road network - Parking As is evident also in the roads examined, most collecting roads of the urban complex of Heraklion, with very few exceptions in areas of the historical centre and in the areas of the new city expansions ISBN: 978-960-6766-87-9 50 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 that "were first built and then planned", do indeed have the necessary geometrical characteristics, with a road surface width ranging from 8.00 to 10.00 m. The local roads bear small traffic loads, are characterised by small lengths and widths and, in many cases, are blind alleys, mostly in the areas of the new expansions, south of the city. Another characteristic of these roads is their intensely varying width, as well as the phenomenon of non-opened sections, something that negatively affects the city's accessibility. As pertains to parking, it is remarked that, with very few exceptions, it takes place on the roadside, either on the one, or on both sides, both in the centre and in the surrounding areas. In some cases, parking is permitted and in some other cases, it is not. Problems related to parking are remarked mainly along the road axes of the historical commercial centre, making the already problematic conditions of traffic even more difficult. Illegal parking includes also the cases of double parking, especially in areas of the city centre, parking on pavements and within small distances from intersections, with or without traffic signs, resulting in reduced road traffic capacity, reduced driver and pedestrian visibility, occupation of pavements and pedestrian crossings, etc. transverse slope, potholes, smooth surfaces); Poor workmanship in the construction of the ramps; Poor workmanship in the construction of the kerbs (unacceptable height); Stairways on pavements; Small size of traffic islands; Insufficient lighting; Non-formed pavements; Big longitudinal slopes; Insufficient pavement surface in relation to the movements (load) of pedestrians. 2.9 Vegetation ΒΛΑΣΤΗΣΗ 19% 49% ΚΑΘΟΛΟΥ ΧΑΜΗΛΗ ΥΨΗΛΗ 32% 2.8 Pavement widths ΠΛΑΤΟΣ ΠΕΖΟΔΡΟΜΙΩΝ Fig. 11: Vegetation In about 20 percent of the roads examined, there is total absence of vegetation, while in 49 percent of the roads, where the vegetation is characterised as high (tall or densely arranged trees), the vegetation of border spaces and balconies has also been taken into consideration (areas with a low bpl and open building pattern private properties). Generally speaking, the trees are absent from most of the city's roads inside the Venetian walls, while the existing trees, especially along roads that have been recently reformed, have been planted recently or are of a small size. 41% 38% <1,20 1,20-2,00 >2,00 21% Fig. 10: Pavement widths Thirty-eight percent (38%) of the roads examined, either lack a pavement or have pavements of small widths (<1.2 m), that do not suffice for two persons simultaneously using the pavement. Pavements with a width of more than 2.0 m are found in areas with the "open on all sides" construction pattern, while pavement widths ranging from 1.2 to 2.0 m are found in areas of the city centre that have been recently reformed. The design and construction problems that hinder the movements of pedestrians are categorised as follows: Poor workmanship in the construction of the pavements (irregularities, poor ISBN: 978-960-6766-87-9 51 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 2.10 Road surface materials and condition ΥΛΙΚΑ ΕΠΙΣΤΡΩΣΗΣ - ΚΑΤΑΣΤΑΣΗ ΟΔΟΣΤΡΩΜΑΤΟΣ 15% 38% ΚΑΚΗ ΚΑΛΗ ΠΟΛΎ ΚΑΛΗ 47% Fig. 12: Road surface materials and condition Generally speaking, in a very small percentage of the roads examined (15%), the condition of the road surface is characterised as poor, mainly due to lack of maintenance, while the in the remaining 85%, the condition is characterised as good or very good. Also, the poor condition of the road surface reflects problems in the underground equipment of the road (water supply and drainage networks, telephone lines, etc.). 2.11 Traffic signs In most part of the length of the roads of the urban complex of Heraklion, there are practically no signs, both for traffic regulation and for informative purposes. This, in turn, affects both the road safety and the intensification of the phenomenon of additional courses and results in the overloading of the road network with vehicles and the increase of noise and air pollution, as well as of energy consumption. neighbourhoods from through traffic and prevent the aimless passage of vehicles through the city centre. Utilisation of appropriate signs (traffic, informative), both horizontal and vertical, in conspicuous places, with the aim to prevent additional courses and to increase the level of road safety, whilst reducing traffic loads. The reduction of the traffic loads and movements into and out-of the city centre can be achieved also with the decentralisation of certain functions (public service locations, shops). The creation of two urban planning centres (i.e. eastern and western), a matter discussed for years, must now be implemented. Replacement and maintenance of underground public utility infrastructures, underground deployment of the now overhead power grids and rebuilding of the road surface, though not necessarily with asphalt. The use of environmentfriendly materials (stone or concrete blocks, stamped and coloured concrete or even stamped and/or coloured asphalt surface), in addition to providing environmental upgrade (light-coloured materials, higher reflectivity), will also improve the aesthetic value of the road [4]. Implementation of the scaled arrangement of the last storeys of buildings (positively affecting sun-lighting, ventilation and dispersal of air pollutants) and of pilotis on the ground floor of new buildings (positively affecting the ventilation of both the buildings and the roads). Installation of technical equipment (wastebaskets and wastebins, lighting elements, etc.) in roads lacking thereof. 3.2 Interventions for pedestrians Widening the existing pavements, so as to enable the convenient crossing of two pedestrians, including the planting of, and the provision of technical equipment for, the pavement (1.50-2.0 m). Discouraging the use of the private car, by reducing the parking places, increasing the cost of parking and imposing more restrictive time limits for parking. Adoption of measures aimed at reducing traffic speeds inside the city and, especially, in local roads [5]. Improving safety, by rebuilding the destroyed parts of pavements and by ensuring anti-slippery surfaces and homogeneity all over the pavement, as well as by ensuring adequate lighting during night hours, without intense changes between light and dark [6]. Creation of pedestrian streets connecting points of historical and cultural interest within the Venetian fortress area of the city and giving 52 ISSN 1790-2769 3 Interventions This paper has presented the parameters that cause problems to the environmental quality of roads in the urban complex of Heraklion. Following this, we present some indicative interventions, which could be undertaken by the state or even private entities, and could, indirectly or directly, help to improve the environmental quality. The interventions proposed relate to the user of the road (pedestrian, cyclist, passenger of public transport means, vehicle driver - even though effort is being put in reducing the latter to the minimum extent possible), focusing on functionality, safety and, of course, convenience and aesthetic value. 3.1 General interventions Classification of the city's road network, with the creation of distinct rings aimed to protect the ISBN: 978-960-6766-87-9 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 priority to pedestrians, at the points of intersection with vehicle traffic roads, through the provision of "green waves" for pedestrians (continuous movement of an average pedestrian with normal walking pace) [7]. Conversion of two-way collecting roads to oneway roads (improvement of road safety through reduction of traffic flow collision points, elimination of head-on collisions and improvement of visibility at intersections). Planting of deciduous trees and bushes on the eastern pavement of roads with a N-S direction and on the southern pavement of roads with a E-W direction, so as to ensure shading during the summer months. 3.3 Interventions for public transport Creation of new routes in areas with poor or no public transport coverage. Covering the needs of the city centre and neighbourhoods with minibuses and operation of transfer stations at the city's gates of entry into the historical centre. Co-ordination of the various routes and transport means, so as to facilitate transfers. Establishment of attractive public transport timetables. Full and precise information at bus stops and inside public transport vehicles for itineraries and other network facilities. Elaborating on the possibility to use environmentfriendly fuels (bio-fuels, gas, etc.). 3.2 Interventions for cyclists The bicycle may be one of the most effective solutions for the problem of inadequate movement and parking areas, in the 21st century city, given that it still guarantees independence of movement. Introducing the use of the bicycle and restructuring the road space are two closely interconnected aims. The one is a prerequisite for the other. Unless the road space is made beautiful, attractive and cyclist-friendly, residents will continue to choose cars or motorcycles, i.e. means of transport that cannot be categorised as environmentally sound, regardless of the classification criteria used. On the other hand, unless the citizens start to use the bicycle for some of their journeys, there will be no suitable awareness, that would urge the authorities to proceed with the implementation of the city's reformation works [4]. Based on the above, bicycle use in the urban complex of Heraklion, an area with favourable longitudinal slopes and climatic conditions, is achievable by: Rebuilding the road surface with the materials mentioned above; Allowing the use of pedestrian zones to cyclists; Construction of bicycle parking areas at the public transport terminals, thus allowing combined transport schemes; Utilisation of buses with bicycle carrying capability; New lines on principal and collecting roads, creating cycle lanes; Replacement of traffic signalling at intersections, ensuring priority for cyclists; Creation of traffic loops in the city's historical centre, with privileged bicycle traffic [8]. 3.4 Public and private initiatives The interventions mentioned above relate mainly to the government, at the local and/or central level. However, without private initiatives, no intervention is eventually feasible. Familiarising with the use of the bicycle or of public transport, or moving on foot, by simultaneously reducing the use of the private car, depends on the citizens, as does also the planting of border spaces in the areas built with the "open on all sides" pattern or of balconies on multi-storey buildings, and the use of environmentally sound materials for developing roadside uses. Finally, it is up to the users of roads to preserve the environmental quality of the roads in the city of Heraklion. 3.5 Effect of interventions on parameters of environmental quality Almost all of the above interventions aim, directly or indirectly, at discouraging the use of the private car in the city's roads and by affecting some of the environmental quality parameters in the city, as shown on the following table: 4 Conclusions It is obvious, both from analysing the current situation of environmental quality in the roads of the urban complex of Heraklion and from the proposals proposed, that there is a need to create a Geographic Information System. The undoubted complexity and interaction of the information issues that are evoked by the modern needs for urban, traffic, spatial arrangement and environmental planning, can easily be addressed with the creation of a database, which could contribute to the qualitative improvement of research, in issues pertaining to environmental design and, simultaneously, allow for a diachronic information and control of the factors making up an environmental policy. 53 ISSN 1790-2769 ISBN: 978-960-6766-87-9 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 S u n li g h t S u n p r o t e c ti o n V e n ti l a ti o n W i n d p r o t e c ti o n N o i s e Ai r po llu tio n V is u a l d is t u r b a n c e Tr aff ic loa ds Us e of ve hic les S a f e t y Aes theti c Val ue Classification of the road network Road signs – Adjustment of road signalling Decentralisation of central functions Network repairs - rebuilding of roads Penthouses –pilotis Installation of equipment Widening of pavements Vehicle parking policies Speed reduction Reconstruction of pavements Pedestrian streets One-way roads Planting New bus routes Minibus Facilitating transfers Denser itineraries Information on bus facilities Table 2: Effect of interventions on parameters of environmental quality As is evident from the collection of the data necessary for this paper, the material that is needed for creating the said database, is indeed available. What remains to be done is the processing, updating and illustration of this material in the form of maps or charts of quantitative data, as well as its continuous updating. References: [1] Zivas, D. The public space of the city. Its formation, function and educative mission. The aesthetics of cities and the policy of intervention – contribution to the regeneration of the urban space, Unification of the Archaeological Sites of Athens S.A., Athens 13-14 October 2003 (in Greek). [2] Aisotos, G. The Greek public space. The aesthetics of cities and the policy of intervention – contribution to the regeneration of the urban space, Unification of the Archaeological Sites of Athens S.A., Athens 13-14 October 2003 (in Greek). [3] Vlastos, Th., Siolas, A. The contribution of transport networks to the strategy for articulating and reuniting urban entities in view of restructuring the city. The case of western Athens, Technika Chronika, vol. 14, no. I, 1994 (in Greek). [4] Vlastos, Th., Polyzos, I. Policies for the urban environment – the European experience. Part A, Chap. 2 in Design, environmental impact and methods for their assessment. Urban planning and environmental impact. Hellenic Open University, Patras, 1999 (in Greek). [5] Racioppi F. Dora C. Krech R. O von Ehrenstein, A physically active life through ISBN: 978-960-6766-87-9 54 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 everyday transport, Word Health Organization, 2002 [6] Siddhartha Sen, Toward a typology of transportation-related urban design problems and solutions: case studies of small and medium sized cities in the eastern United States, National Transportation Center , Morgan State University Baltimore, Maryland, 1999 [7] Gehl Ian, Public spaces and public life, Adelaide , 2002 [8] Dekoster, J., Schollaert, U. (1999). Cycling: the way ahead for towns and cities. Office for Official Publications of the European Communities, Luxembourg, 1999. ISBN: 978-960-6766-87-9 55 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 Theoretical Aspects Concerning the Adoption of the Number and Dimensions of Brake Discs for Railway Vehicles CĂTĂLIN CRUCEANU, CAMIL CRĂCIUN Faculty of Transports, Rolling Stock Engineering Department POLITEHNICA University of Bucharest 313 Splaiul Independentei, sect. 6, 77206, Bucharest ROMANIA c_cruceanu@yahoo.com, http://www.pub.ro Abstract: In order to ensure the safety of the railway traffic, the braking system of the vehicles is very important, especially when running with high speeds. In our study there are pointed out some particular aspects that have to take in consideration for a rational and optimum adoption of the number and dimensions of the brake discs that equip a railway passenger vehicle: the maximum developed temperature at the brake pads - disc interface, the disc’s resistance regarding the thermal fatigue and the vertical dynamic forces due to the unsprung mass. A calculus case study is also developed and analyzed, in order to emphasize the theoretical aspects presented in the paper. Keywords: Safety of traffic, Brake discs, Temperature, Thermal fatigue, Vertical dynamic forces. 1 Introduction Generally, the braking force has to assure the safety of the traffic, which for the railway vehicles mainly means to ensure a certain braking capacity, respecting also some essential aspects like: not blocking the wheels during braking, respecting the imposed braking distance and the limitation of deceleration to acceptable values for the passengers’ comfort. While designing a disc braking system, considering the necessary force value known, the main factors that determine the choosing of brake discs number and their dimensions are: the maximum developed temperature at the brake pads - disc interface, the disc’s resistance regarding the thermal fatigue and the vertical dynamic loads due to the unsuspended mass, which obviously increases with the number of the brake discs that endow the vehicle. During the exploitation of disc brake endowed with composite material brake pads in respect to [1], the temperatures must not exceed 350…375oC in order to avoid a fast and irrecoverable diminish of braking capacity. That is why the main criterion in choosing the type, the geometrical dimension and the necessary number of brake discs is connected with the maximum reached temperature values during an emergency braking action from the maximum constructive vehicle’s speed. Because brake discs are subjected to successive heating and cooling processes, the disc’s crown is supplementary exposed to thermal fatigues. So, when choosing the brake discs number and dimensions, it is important to take into consideration the number of braking-loosening cycles until bracks may begin to emerge. The criterion is more growing in importance considering the fact that in the case of no sectioned brake disc usage, their replacement technologically implies also to remove the wheels away from the wheel axle. From this point of view, the ideal solution is when reaching the limits of the brake disc resistance at thermal fatigue coincide with the wheels wearing out limits. This fact makes the brake discs replacement to come together with the removing of the wheels from axels, avoiding supplementary costs due to vehicles withdraw from service. 2 Theoretical bases 2.1 Thermal calculation The friction surfaces temperature during braking process represents a determinant criterion for wearout and lifetime of the friction coupling. The composite material of the brake pads reaches thermal limits around 375oC. If exceeded, a fast and even decisive brake pads wear-out occurs, along with unpleasant smells. The situations when the maximum thermal charges of brake disc are expected are: the case of ISBN: 978-960-6766-87-9 56 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 emergency braking action from the vehicle’s maximum constructive speed and the case of service braking to maintain a constant running speed while descending long and great declivities. In our study, we considered only the first case because during this type of action, not only the highest temperatures are reached within the pads - disc interface, but also occurs the largest temperature difference between the external friction surfaces of the disc and the central auto vented area. When two bodies slide with respect to each other, the kinetic energy is primary transformed into deformation mechanical work and it is to be recovered after the deformation is completed by means of heat within the sliding and separating parts due to the friction. The heat quantity taken by the aforementioned ones is insignificantly small; therefore it was neglected within our study. In the case of disc brake, the effective friction processes between the pads and the disc are extremely complex due to the fact that the present time brake pads, due to their composite structure [1], do not have constant chemical-physic proprieties, the organic contained elements being subject of a series of transformations under the influence of temperature increase. The heat distribution between the brake disc and the friction pads is mostly dependent on material characteristics, among whom a major influence is due to the density ρd,g [kg/m3], the thermal conductivity λd,g [W/m.K] and the specific heat cd,g [J/kg.K] of disc’s (index d) and braking pad’s materials respectively (index g). Denoting Qd and Qg [J] the heat quantities assumed by the disc and the braking pads respectively, one could be expressed in the following manner [2]: ρ d ⋅ λ d ⋅ cd Qd = . (1) Qg ρ g ⋅ λ g ⋅ cg Because the braking disc is not entirely covered by the friction pads, within computing we have to consider the ratio between the disc surface Sd and the pads surface Sg. Denoting the ratio of heat’s division between the disk and pads with: ρ d ⋅ λ d ⋅ cd S d ϕ S (2) ⋅ ϕc = d ⋅ d = ρ g ⋅ λ g ⋅ cg S g ϕg S g and considering Q [J] the heat quantity generated during the friction process, the heat quantities assumed by the pads and by the disc are: 1 Qg = Q ⋅ (3) 1 + ϕc ϕ Qd = Q ⋅ c . (4) 1 + ϕc The brake disk assumes the most part of the heat, usually more than 90% [3], through the effective contact surface of the friction coupling. From the brake disc friction surface, the heat is conductively transmitted to its crown. Together with the increase of the external surfaces temperature, due to the relative displacement of air with respect to them, an important convective thermal transfer process is developed, depending on the instantaneous speed and accentuated due to the auto-ventilated construction of the disc. Once with the temperature’s increase, a radiant giving up transfer of heat phenomenon appears. The difference between the received heat quantity and the evacuated one from the brake disc, within its different ways and zones, leads to the development of a temperature field that can be specified by the means of integrating the thermal equation: d (m ⋅ c ⋅ dθ ) dq p = dqcond + dqconv + dqrad + , (5) dt where dqp, dqcond, dqconv, dqrad [W] are the thermal flux received by the disc during braking, the conductive thermal flux within the metal mass of the disc, the convective and radiation like given up thermal fluxes, m [kg] and c [J/kg.K] the mass and the specific heat of the material the disc is made of, dθ [K] the variation in temperature and t[s] the duration of studied phenomena. In the description of equations defining the thermal phenomena of brake’s disc, the main initial hypothesis adopted are [3]: - the mechanical work of the friction forces are entirely transformed into heat that is assumed by the friction force coupling; - the heat distribution that resulted the friction is uniformly spread on the disc surface; - the materials the brake discs are made of are considered to be homogenous; - the thermal charge is symmetrical; - the conductive transmission of heat between the brake disc crown and the hub is considered to be not significant; - the physical properties of disc and pad materials are not temperature depending, with the exception of the conductive thermal transfer coefficient and the specific heat, by whom the temperature variation is considered within computing; - the conductive thermal transfer is unidirectional within the disc crown, from friction surfaces to the central ventilation channels; - the convective thermal transfer is the most important phenomenon, concerning a surface equal with the total disc’s surface, excepting the friction ISBN: 978-960-6766-87-9 57 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 surfaces pads’ sizes and it is larger in the inner side of the disc, due to the ventilation fans; - the radiant thermal transfer might be neglected. Due to these facts, relation (5) can be written as [3]: dT dQd = −λ ⋅ Scond ⋅ ⋅ dt + S conv ⋅ α ⋅ dT + dx + ρ ⋅ c p ⋅ dV ⋅ dT (6) Regarding the transitional character of the phenomenon, for the analytic integration of the thermal equation, the finite differences method is used with a corresponding selection of brake disc and time step t parameterization. 2.2 The brake disc endurance to thermal fatigue The effort state within the disc is as much more complex as larger the difference in quantity and gradient of the two thermal fluxes is, one proportional with the heating quantity onto the active sides of disc, the other one caused by the exhaust of heat by means of conduction, convection and radiating. Within a full braking – releasing cycle, the most important aspect is the increase with higher gradient of temperature onto the active sides, starting from the moment when the friction forces are applied. The disc side’s friction temperature rises until a maximum value, corresponding to the moment of reaching an equilibrium between the receiving and the exhausting of heat, afterwards the temperature shows a negative gradient until the interruption, during the interruption and then at the replay until the next braking action. The main thermal flux is perpendicular onto the friction surface plane, in the direction of the interior ventilation channel. Therefore an axial temperature gradient rises, to which a tensional state will correspond, that depends on the value of the gradient, of the material elastic characteristics and on its effort state with respect to linear dilatation. For the crown built in one piece, the following efforts appear [4]: E σ=− ⋅ α ⋅ Δθ , (7) 1− ν where E is the longitudinal elasticity module of disc’s material, ν is Poisson’s coefficient and α is the coefficient of linear thermal dilatation, the temperature difference Δθ between the disc friction surface and the inner side of it being also influenced by the value of thermal conductivity and material specific heat, parameters dependent of local state of temperature. The value of effort σ is negative (compression) for Δθ positive (the friction surface temperature greater than the internal one) and positive (traction) for Δθ negative; this will happen in a single cycle braking – release. When repeating such cycles, together with the superposition onto the other existing disc tensions, which may lead to plastic breaking of material, thermal fatigue phenomena appear. This is the reason of occurring superficial micro fissures, mainly in radial way (orthogonal on effort σ), fissures that, in time, can contribute to the damaging of crown, so imposing the necessity of being replaced before reaching its wear out limits. Regarding the studies and experiments, the O.R.E. B126 Commission, within [5], taking into account as unique parameter the maximum temperature reached within the brake disc, computed the curve of dependence of the limit braking cycle number until the occurrence of fissures in the disc crown. 1000 θmax [0C] 100 10 10 102 103 104 105 106 Fig. 1. The number of braking – releasing cycles until the occurrence of thermal fissures. This Wöhler type curve is presented in fig. 1. In order to use this information within a calculus program in an efficient manner, we established the mathematic dependency of the number of limit cycles nL function of maximum temperature θmax, so: - for θ < 1490 C : nL = 6,164 ⋅ 108 ⋅ e −0,06⋅θ max (8) - for 1490 C ≤ θ < 2000 C : nL = (0,16 ⋅ θmax − 23,32 ) ⋅ 10 4 - for θ > 200 C : nL = 1,57 ⋅106 ⋅ e −0,024⋅θ max 0 (9) (10) 2.3 Vertical dynamic wheel-rail contact forces Due to the elastic and damping characteristics of the track and also due to its vertical irregularities, at the contact between wheels and the rails, there develop certain vertical dynamical forces that load and discharge successively, supplementary, the wheel ISBN: 978-960-6766-87-9 58 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 sets. Because the values of the natural frequency of the wheel – rail system are approximately one order higher than the ones of the sprung masses of the vehicle, for the theoretical study, the two oscillating systems might be considered separated. These aspects were theoretically studied and presented in [6,7], considering that the longitudinal leveling irregularities of the two rails of the track are in the same phase. Within that study there have been established the computing relations of dynamical contact forces, the mathematical model taking into account the random character of excitations due to geometric vertical irregularities of the track. mn 3 Case study Based on the aforementioned theoretical elements, a computer packet of programs has been realized in order to set the temperatures in the brake disc. Furthermore, we extended these to determine also the limit number of cycles of braking – release the disc can stand and the vertical dynamical wheel – rail contact forces. We studied the case of a first class passenger vehicle with 10 compartments, each having 6 sitting places. The mass of the unloaded vehicle was considered to be 40 t. The study was performed for a running speed in the domain 160…350 km/h, taking into consideration the fact that each wheel set is endowed with 1…4 braking disks situated on the axel body. For each situation, only the disc brake was of interest, considering that the maximum braking force is obtained only with this type of braking system. θ [ oC ] 1500 z η cc ρc 1000 a. brake disc of 640 mm diameter Fig. 2. Model for unsuspended vehicle’s mass vibrations. 500 Starting from the mechanical model presented in fig. 2, the movement equation of the unsprung mass mn is [6,7]: •• ⎛• •⎞ (11) mn ⋅ z + ρc ⋅ ⎜ z − η ⎟ + cc ⋅ (z − η) = 0 , ⎝ ⎠ where z is the displacement of unsprung mass, η(t) the excitation due to railway’s vertical irregularities and cc and ρc the stiffness, respectively the track’s shock absorption factor. Considering η a harmonic function, using complex notations, function of the system’s natural pulsation ωc = cc mn and 0 1500 1000 b . brake disc of 610 mm diameter 500 1600 1200 800 400 0 ω = 2 ⋅ π ⋅ V (3.6 ⋅ L ) being the excitation pulsation, the transfer module function in vertical amplitude is [6,7]: ⋅ (ω ωc ) c The response factor of dynamical vertical charge is: (13) H Δq (ω) = mn ⋅ ω2 ⋅ H ηz (ω) 2 2 Dc = ρc 2 ⋅ cc ⋅ mn the damping ratio of the track, ( ) c . brake disc of 590 mm diameter H ηz (ω) = [1 − (ω ω ) ] + 1 + 4 ⋅ Dc2 ⋅ (ω ωc ) 4 ⋅ Dc2 2 2 (12) 0 1 2 3 number of discs / wheel set 4 Within the previous relations, V [km/h] is the running speed of the vehicle and L [m] is the wave length of the vertical irregularities of the track. Fig. 3. The maximum reached temperature within the disc at the emergency brake action from a maximum running speed of: 350 km/h; 300 km/h; 250 km/h; 200 km/h; 160 km/h; limit of 350oC. ISBN: 978-960-6766-87-9 59 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 The evolution of maximum temperatures within the braking disc, function of the number and diameter of each disc on the vehicle axle and their dependence upon the braking start speed is presented within the diagrams in fig. 3. On the diagrams the maximum admitted temperature of 375oC was set, for determining the exact necessary number of brake discs mounted on an axle, function of their diameter, on the base of temperature criterion. In the case of using 640 mm diameter brake discs (see fig. 3, a) it is enough that an axle is provided with two of them for a maximum running speeds of 250 km/h. For higher level speeds like 300 km/h, the axle should be endowed with three brake discs, and even four in the case of running speeds exceeding 300 km/h (300…350 km/h). In the case of using 610 mm diameter brake discs (see fig. 3, b), four discs on axle are needed for a speed of 300 km/h and three for a speed of 250 km/h. The braking discs having the diameter of 590 mm (see fig. 3, c) cannot be used for speeds of 350 km/h, and a speed of 300 km/h necessitates the mounting of four discs onto one axle. Furthermore for the speed of 250 km/h there are necessary three discs onto each axle. In the situation of emergency braking it can be observed that, regardless of diameter size, the use of two braking discs instead of one mounted on the axle leads to a diminish of approximately 52% of the maximum temperature. The upgrading from two to three brake discs per axle leads to a diminish of approximately 35% of maximum temperature, while the mounting on the axle of four brake discs instead of three ensures a drop in the maximum reached temperature of approximately 27%. Concerning the fatigue resistance, within the diagrams in fig. 4 it is presented the number of braking – releasing cycles nL the brake disc can resist function of the number and diameter of the mounted discs on every vehicle axle and their dependence on the start braking running speed. Statistics reveal that in generally, within the interval of two planned reparations, the vehicle does between (5…10) .104 brake – releasing cycles. From the analysis of obtained results it is shown that on the studied speed domain, in the case of mounting onto every axle of one or two brake discs, these would resist 34…5300 brake – release cycles, that number being insufficient. Three brake discs having the diameter of 640 mm onto each axle could be used for a maximum velocity of 300 km/h, while the discs having the diameters 610 mm or 590 mm could be used for a maximum speed of 250 km/h. In the situation of mounting four brake discs on every axle, the 640 mm diameter ones could be used at speeds of 350 km/h and the ones with smaller diameter at speeds of 300 km/h. It is made the observation that the fig. 4 results are referring to emergency braking cases, while in normal operating are usually used graded service braking actions and in the case of high speed rolling, the friction brakes are used just in the case in which dynamic braking systems do not ensure the necessary braking power. That is why the maximum reached temperatures within the brake discs are substantially smaller, furthermore greater time durability at such brake – releasing cycles. nL 3 . 103 2 . 103 1 . 103 a. 1 brake disc / wheel set 6 . 103 4 . 103 0 b. 2 brake discs / wheel set 2 . 103 105 8 . 104 6 . 104 4 10 . 4 0 c. 3 brake discs / wheel set 2 . 104 0 6 . 105 4 . 105 2 . 105 d. 4 brake discs / wheel set 0 200 240 280 160 320 running speed at beginning of braking [km/h] Fig. 4. The number of cycles brake – release nL the disc can resist, function of the brake action starting speed and the disc diameter: 640mm; 610mm; 590mm. ISBN: 978-960-6766-87-9 60 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 The evolution of dynamical vertical forces at the wheels-rails contact, function of the number and diameter of discs mounted onto every vehicle’s axle and their dependence on the running speeds is shown in the diagram in fig. 5. As expected, it is obvious that the dynamical vertical loads increases with the running speed. For the studied case, at speeds of 350 km/h, there represent approximately one third from the vertical load magnitude onto one wheel. Concerning the influence onto the dimensions of brake discs, this is approximately 14% from the size of the vertical static load. ΔQ [kN] Φ 640 mm 20 Φ 610 mm 15 Φ 590 mm domain 250…310 km/h three discs being sufficient, regardless of their sizes, while over 300 km/h four braking disks become a necessity; - when choosing criterion for the dimensions and disc numbers there has to be recorded the interval between planned reparations in conformity with the type of vehicle, so the number of braking – releasing cycles the disc supports to be in concordance with these reparations; - finally, the process of adopting in ended regarding the safety of traffic dependent to the ratio (Y Q ) , where Q represents the vertical component of the effective reaction of the rail at the derail limit dependent on the guidance force Y, taking into consideration the fact that the dynamical tasks supplementary charge and discharge the axles with the corresponding value. The pointed out aspects, the elaborated computer program packet, as the indicated interpretation and analysis mode of the obtained results consists of an determination method of necessary number and dimensions of brake discs for endowing a vehicle that will run safety with a maximum imposed speed. References: [1] Fiche U.I.C. 541-3: FREIN – Frein à disques et garnitures de frein à disques, 4e édition, 01.07.1993. [2] Saumweber, E., Temperaturberechnung in Bremsscheiben für ein beliebiges Fahrprogramm, Leichtbau der Verkehrsfahrzeuge, Heft 3/1969, Augsburg. [3] Cruceanu, C., Frâne pentru vehicule feroviare (Brakes for railway vehicles), Ed. MATRIXROM, Bucureşti, 2007, ISBN 978-973-755-200-6, 388 pag. [4] Camposano, P., Poli, A., Attitude dei dischi frenoferroviari alle più severe prestazioni, Ingegneria ferroviaria, nr. 5 – Maggio, 1990, p. 357367. [5] O.R.E., Quéstion B 126. Problèmes de freinage: Rapport no. 9, Utrecht, 1988. [6] Sebeşan, I., Dinamica vehiculelor de cale ferată (Dynamics of railway vehicles), Ed. Tehnică, Bucureşti, 1995, ISBN 973-31-0887-1, 317 pag. [7] Sebeşan, I., Hanganu, D., Proiectarea suspensiilor pentru vehicule pe şine (Casting of suspensions for railway vehicles), Ed. Tehnică, Bucureşti, 1993, ISBN 973-31-0451-5, 244 pag. 10 1 5 150 200 250 300 V [km/h] Fig. 5. Vertical dynamical forces in the case of four brake discs mounted onto an axle (1- one disc with 590 mm diameter mounted onto the axle). 4 Conclusion The main conclusions that result from the study are: - in the case of mounting two brake discs on every axle, the discs having a diameter of 640 mm can be used at a maximum speed of 250 km/h, the ones having 610 mm at 210 km/h and the ones having 590 mm at a maximum speed of 180 km/h; - regarding the number of necessary brake discs mounted onto an axle, for the maximum velocity of 160…180 km/h two brake discs are sufficient regardless of their diameter, for the maximum speed between 180…250 km/h two 640 mm diameter or three 610 mm diameter discs can be used; when over passing a running speed of 250 km/h, there is no longer sufficient to use only two discs so, on the ISBN: 978-960-6766-87-9 61 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 Evaluation of the Sustainable Ecological Pond Management in Campus, Miaoli, Taiwan KUEI-YANG WU, WANN-MING WEY, CHI OU-YANG, YA CHUNG, LIN Department of Architecture, National United University Miaoli, TAIWAN 36017, R.O.C. Abstract: This study aims to promote sustainable ecological pond management in campus for rural area by integrated three high levels (ecosystem condition, soil and water, and management planning) to provide school students and residents realizing the effects on environment of ecological pond. In this research we first reviewed literatures and then interviewed with experts by questionnaire using AHP method. Results show that there are three themes and associated benchmarks in this study: (1) ecosystem condition: site and layout, natural environment, cultural heritage, traffic conditions, existing biology, groundwater system; (2)soil and water: water in pond, infiltrated in pond bottom, soil type and environment benefit to ecology, plant in water, biodiversity, purify water quality by microbiology, and(3)management planning: safety facility, economy management, regular maintenance, health and sanitation, personnel quality, environmental education. These three themes expressed by AHP questionnaire data in sustainable ecological pond management in campus, Malio, Taiwan can be ranked and weighted collectively for 32.79%, 20.33% and 46.88 %, respectively. The method may be helpful to landscape architect, environmental education experts and decision-maker in government for sustainable water environment planning. Keywords: FAHP questionnaire, Ecological pond, Soil and water, Groundwater, Management planning. 1. Introduction The ecological pond in campus is a good example of many interacting systems, with constant changes in the ecology and a stable overall effect on the environment, also function as an environmental education [4]. The system balance and stability is only disturbed when we overload the pond with pollutants or nutrients from human systems. Species diversity in ecological pond can reinforce the stability of these systems with redundancy in function. The result is self-organization, self-repair, self-reproduction, and a great ability to adapt to perturbations in external conditions. If campus processes can have these same positive attributes, they will be highly effective and often restorative. In recent time potentially increasing risk comes from water shortage, and the impact from global change [9] and local urban heat island seriousness [2, 5, 7] that both make improve or mitigate these problems, such as by building ecological pond, being urgent and important issues in Taiwan. For assessing ecological pond management strategy in rural campus we adopted the analytic hierarchy process (referred 1 ISBN: 978-960-6766-87-9 62 as AHP) that were applied in diverse fields by these authors [10, 13-15] using questionnaire survey to obtain important management levels. To instill our campus with sustainability in Miaoli, Taiwan, we have scrutinized the school’s local environmental characteristics, cultural background, coordinated with the school’s existing layout criteria and conditions, in anticipation of systematically integrate school resources. Many elementary schools in Miaoli, Taiwan aim to attain sustainable campus that include the Shui Wei elementary school where is situated in Puli Township, Nantou county, and is blessed by a natural ecology setting and rare natural beauty. Recent a nearby irrigating water body was routed to the campus, which not only serves to enhance the unique beauty of its campus but also helps to culminate a nature science classroom for all students. Among them, water recycling and planning ecological pond is one important concept in the sustainable partial campus reengineer plan proposed by the school pertains to an “Ecological cycle revamp project”, that need to be fully integrated to make up a recyclable ecological campus system in order to meet the rudimentary concept in resource conservation ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 and resource recycling and support creating a campus with effective utilization of water resources, coordinated with showcasing the school’s emblematic growth of the large leaf mahogany, helping to excel greater material recycling and reuse. The challenge for ecological engineering is to take the mechanisms, pathways, nutrient flows and organisms found in all of these systems and design them into mesocosms that are the ecological technologies of the future. In this scenario a waste treatment facility is no longer a sewage treatment plant, if we use a nutrient and materials management system, or a water-based pond with useful products and a environmental conservation as well as education as central to the design criteria. Lin et al. [4] pointed out in order to promote environmental education for urban residents by integrated three ponds planning to provide visitors realizing the effects on environment by sightseeing process. Their results show that there are four themes in integrated three ponds planning: (1) resource recycling and wetland purification; (2) infiltration and hydrological circulation; (3) biodiversity, urban microclimate, and environmental education; and (4) aesthetic feeling of landscape: landscape beauty. Wang et al. [12] thought wastewater and storm water management are incorporated into ecological design by open water channel, storage pond, and constructed wetlands to form green water-ways or ecological corridors, which connect remnant patches into an ecological network within the urban area and lead to an integrated water and ecological management. Kim, et al. [3] present results from a case study which examined the habitat characteristics and the macrophyte and macroinvertebrate biodiversity of eight farm ponds on four properties in the Stanley Catchment, Southeast Queensland, Australia. Ou, et al. [8] pointed out a constructed wetland-pond system consisting of two free-water-surface-flow (FWS) wetland cells, a scenic pond, and a slag filter in series was used for reclamation of septic tank effluent from a campus building. To promote environmental education for ecological pond in rural campus by providing campus students and local residents in realizing 2 ISBN: 978-960-6766-87-9 63 the effects on environment water resource conversation process, this study is focused on: (1) reviewing literatures and then interviewing with experts by questionnaire using AHP method; (2) integrating an ecological pond assessment strategy; and (3) demonstrating its application in rural casmpus. 2. Site and Survey Our study aim to investigate three elementary school campus selected from HiKo elementary school, ChuNan elementary school, located at northern Miaoli, Taiwan, ChungKon elementary school at middle Maoli; KaChan elementary school, YuanLi, at southern Miaoli. Fig. 1 shows the campus locations of sustainable ecological pond management in campus, Miaoli, Taiwan. After surveying these campus and its dwellers, we found two probable main problems when promoted the campus’s ecological pond: (1) not long term plan for ecological pond, and (2) to select site satisfied an ecosystem condition to built ecological pond. 3. Method In order to obtain the assessment criteria for the ecological pond in a campus our research methods contained the literature analysis, expert opinions method, questionnaire survey and Fuzzy Analytic Hierarchy Process (referred as FAHP). Firstly, we summarized the detailed descriptions related to principles and elements of the eco-pond planning from each literature proposed respectively by Pong [9], Tsai [11], etc., to acquire a primary hierarchy structure of assessment. After confirming the hierarchy structure of assessment, we implemented AHP questionnaire survey to understand the experts’ opinions about the comparison of each pair benchmark (variable) through the questionnaire method. Then we utilized the FAHP to analyze results in the questionnaire survey. 3.1 Fuzzy Analytic Hierarchy Process Actually, it existed the defect of vagueness upon the decision-making of the analytic hierarchy process. Therefore, we adopt the FAHP to establish the assessment criterion for ecological pond in the campus. Also, we combined the fuzzy theory and AHP to make up the imperfection of AHP. The ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 FAHP is based on the conventional theories of AHP, thus we adopted fuzzy numbers to represent the inaccurate values by originally presuming, and then to perform a decision analysis using the fuzzy number algorithm in decision rules. Overall, this method of FAHP is consisted of the fuzzy linguistics, defuzzy and normalization. Its operation steps are as follows (Fig. 3). 3.2 Hierarchy structure for assessment After we compiled various literatures, there are three high level aspects in the hierarchy structure for assessment. It is "ecosystem condition", "soil and water", and "management planning". Under three aspects, we established total 19 assessment criteria in this study. Among them, the assessment criteria under the "ecosystem condition" aspect including “site and layout”, “natural environment”, “cultural heritage”, “traffic conditions”, “existing biology”, and “groundwater system”. The assessment criteria under the soil and water aspect included “water in pond”, “infiltrated in pond bottom”, “soil type” and “environment benefit to ecology”, “plant in water”, “biodiversity”, “purifying water quality by microbiology”. The assessment criteria under the “management planning“ aspect included “safety facility”, “economy management”, “regular maintenance”, “health and sanitation”, “personnel quality”, “environmental education”. Every criteria (variable) is described in Table 1. 3.2.1 Ecosystem Condition aspect The content of ecosystem condition aspect is comprised of the natural and present environmental condition. Therefore, this aspect includes “site and layout”, “natural environment”, “cultural heritage”, “traffic conditions”, “existing biology”, and “groundwater system”. 3.2.2 Soil and Water aspect The “soil and water” aspect indicates the construction for the pond itself. That comprises all necessary facilities, such as “water in pond”, “infiltrated in pond bottom”, “soil type” and “environment benefit to ecology”, “plant in water”, “biodiversity”, “purifying water quality 3 ISBN: 978-960-6766-87-9 64 by microbiology”. 3.2.3 Management Planning aspect The “management planning” aspect means the personnel management. That is comprised of “safety facility”, “economy management”, “regular maintenance”, “health and sanitation”, “personnel quality”, and “environmental education”. 3.3 The assessment criteria for the ecological pond in a campus Based on Buckly’s (1985) FAHP we combined with Triangular Fuzzy Number, COG Method of FAHP proposed. Due to the correction value is a result of its aspect divided by total numbers of assessment criteria. Firstly we multiplied the total weight of each assessment criterion and the correction value, then sorted the priority order for the ecological ponds in campus with corrected weight values. Its correction value equals a value that the criterion number (6) divided by total numbers of criterion (19). The correction value equals 6/19. Therefore, the correction value of base environment equals a value made by the total weight and the correction value multiply (0.0647 × (6/19) = 0.0204). The calculation result is shown in Table 1. Besides we will adopt geometric means to integrate the experts’ opinions and calculated the related fuzzy weight value. Eventually we will perform the defuzzification and hierarchy combination processes by means of above results, and then corrects the relative weight among assessment criteria. The correction value is a result of its aspect divided by total numbers of assessment criteria. Thus, prior to the sorting process for the importance level of all indices, it firstly needs to multiply the total weight of each assessment criterion and the correction value, then, sorting the priority order for the ecological pond in campus with corrected weight values that will be utilized upon the consideration and comparison of total importance level of each criterion. For example, there is a “site and layout” under the “ecosystem condition” aspect. These steps are described as follows: (1) establishment of fuzzy positive reciprocal matrix; (2) group integration; (3) calculation of fuzzy weight value; (4) defuzzification; and (5)hierarchy combination; and (6)correction value and the ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 ranking of assessment criteria 4. Results and analysis We received 10 electronic questionnaires via 10 e-mails to experts. The consistency ratio (C.R.） are less than 0.1 after we performed testing the reliability and validity analysis. Therefore all questionnaires are considered as valid questionnaires in this ecological pond evaluation research. 4.1 Results Analysis of weight calculation in three aspects Shown in Table 1 it is able to find out the weight value of three aspects through the calculation results of fuzzy numbers, group integration, fuzzy weight, and defuzzification etc. The sort sequence would be the "management planning” aspect (0.4688), "ecosystem condition” aspect (0.3279), and "soil and water” aspect (0.2033). According to the calculation results, it show that the most important aspect is the "management planning” aspect in the construction of ecological pond in campus, followed as "ecosystem condition” aspect, and the last one is In term of the "soil and water” aspect. “sustainability” it was expected that it can maintain the sustainable environmental management through a long-term project. Based on this concept, the present study will confirm that the management planning of ecological ponds constructed in the campus is the most important aspect. Additionally, the ecological pond is an artificial construction, because it is needed a longer time to achieve a better adaptation for a circumstance of mutualism in this pond and its surrounding environment. Overall, the construction of an ecological pond not also satisfies a current need but also attains the long term environmental balance, i.e., the ecological pond requires building the good concept toward a sustainable management attitude. 4.2 Results Analysis of criteria weight “Ecosystem condition” aspect will be considered as basic and important condition that can be satisfied to build ecological pond by managers. Hence, all creatures living in this environment will be maintained by a self-sufficiency living 4 ISBN: 978-960-6766-87-9 65 circumstance under a condition of mutual-aid, such as soil and water supplied by in-site underground, climate etc. Results of calculation for the most important assessment criterion under "ecosystem environment" aspect is “groundwater system” (0.0282). Under the “soil and water” aspect, the" Infiltrated in pond bottom" (0.0129) is the most important variable, which means the mater of bottom itself has also the primary element in planning the ecological pond in a campus, followed as the benchmark “plant in water”. That can absorbed and filter the polluted water in this pond, thus it is the second critical variable in this aspect assessment. Under the “management planning” aspect, the "regular maintenance" (0.0327) is the most important variable. It is well known that maintain a pond is the most basic task upon the construction of such pond. Especially, the concept of sustainable development has got higher concerned in our campus. Consequently it is necessary to have insight in building ecological pond with a macro and long-term viewpoint. Overall, the life of a pond is totally relied on the regular and efficient maintenance (Table 1). 4.3 Calculation Results of 19 assessment criteria Among those 19 assessment criteria, the" regular maintenance" (0.0327) has the maximal ratio, the followed as " safety facility" (0.0311), " groundwater system" (0.0282), " environmental education" (0.0277), "natural environment" (0.0276), ". personnel quality" (0.0221), " water in pond " (0.0207), " site and layout" (0.0204), " health and sanitation " (0.0183), and " economy management " (0.0161). Under the “management planning” aspect, the sequence of safety, economy, maintenance, health and sanitation, personnel, plans etc. are mostly occupying in the first-half ranking of total numbers of assessment criteria. If we compare results in three aspects, the management aspect is the most important variable which demonstrates that the management aspect is the most critical variable. 5. Conclusion This study focus on summarizing the principles ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 of planning design for ecological pond constructed in a campus based on the integration of related literatures, and then constructed the hierarchy structure through the experts’ questionnaires. Results show that there are three themes in this study: (1) ecosystem condition; (2)soil and water, and(3)management planning. These three themes expressed by AHP questionnaire data in sustainable ecological pond management in campus, Malio, Taiwan can be ranked and weighted collectively for 32.79%, 20.33% and 46.88 %, respectively. In addition, the limiting factors in designing ecological pond include: (1) rural residents lacking of knowledge and consciousness in water cycling; (2) very few ecological ponds in urban area; and (3) no actively encouragement by local government. Moreover, we further envision a greater purpose of promoting the ecological conservation concept to the community resolving around the school, and to share the sustainable campus management concept with peer schools. Acknowledgement The authors thank NSC (National Science Council) 96-2415-H-239-002-SS3, Taiwan, for financially supporting this research. References: [1]Buckley J.J., Fuzzy Hierarchical Analysis, Fuzzy Sets and Systems17, 1985, pp.233~247. [2]Hsu, HH., Chen, CT., Observed and Projected Climate Change in Taiwan, Meteorol. Atmos. Phys. 79, 2002, pp.87-104. [3]Kim A. Markwell A Christine S. Fellows, Habitat and Biodiversity of On-Farm Water Storages: A Case Study in Southeast Queensland, Australia, Environmental Managemen 41, 2008, pp.234–249 [4]Lin, W.Z., Wang, CH., Huang, CH., Wu, KY., and Chen, SW., Evaluation of the Sustainable Urban Ecological Pond Management in Taipei, Taiwan, WSEAS Transactions on Biology and Biomedicine 7(3), 2006, pp.524-531. [5]Lin, WZ., Tsai, HC., Wang, CH., Chou, C., The Characteristics and Estimation of the Urban Heat Island of a Subtropical City in Taiwan, IASME Transactions 3(2), 2005, pp. 5 ISBN: 978-960-6766-87-9 66 336-345. [6]Liu, SC., Lin, CY., Shiu, CJ., Wang, CH., Liu, GR., Mega-Suburb Heat-Island Effect over Western Plain, BAQ 16 Dec. 2002-18. ,Hong Kong SAR. [7]Oke, TR., The Energetic Basis of the Urban Heat Island, Q. J. R. Met. Soc. 108, 1982, pp. 1-24. [8]Ou, WS., Lin, YF., Jing, SR., Lin, HT., Performance of a Constructed Wetland-Pond System for Treatment and Reuse of Wastewater from Campus Buildings, Water Environment Research 78(12), 2006, pp.2369-2376. [9]Pong, KD., How to Build up a Vital Eco-Pond, Taiwan, Nature Conservation 35, 2001, pp.6~10. [10]Stirn, LZ., Integrating the Fuzzy Analytic Hierarchy Process with Dynamic Programming Approach for Determining the Optimal Forest Management Decisions, Ecological Modelling 194 (1-3), 2006, pp.296-305. [11]Tsai, HN., How to Build an Ecological Pond DIY, Chi-Hsin, Agricultural Development Foundation, Taiwan, 2001. [12]Wang, L., Wang, WD., Gong, ZG., Liu, YL., Zhang,. JJ., Integrated Management of Water and Ecology in the Urban Area of Laoshan District, Qingdao, China, Ecological Engineering 27 (2), 2006, pp.79-83. [13]Wu, KY., Wey, WM., Lin, WZ., Using Fuzzy Analytic Hierarchy Process Evaluates Sustainable Community Management in Miaoli city of Taiwan, WSEAS Transactions on Environment and Development 6(2), 2006, pp.792~799. [14]Wu, K.Y., Wey, W.M., Lin, W.Z., Ecological Preservative Management Evaluation Using Multi-Stage Decision and FAHP Process in Mountainous Maoli, Taiwan. WSEAS Transactions on Environment and Development (accept) , 2008. [15]Wu, KY., Wey, WM., Lin, W.Z., Huang, CH., Disaster Prevention Strategy of Mountainous Community Using Fuzzy Analytic Hierarchy Process in Miaoli, Taiwan, WSEAS Transactions on Biology and Biomedicine 7(3), 2007, pp.538-545. ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 26 China Site and Layout Natural Environment 25 Latitude 24 Taiwan 23 Datong Miaoli County Cultural Heritage Ecosystem Condition Traffic Conditions Existing Biology Groundwater System Water in Pond Infiltrated in Pond Bottom Soil Type and Environment Soil and Water Benefit to Ecology Plant in Water Biodiversity Purifying Water Quality by Microbiology Safety Facility 22 119 120 121 122 Longitude Fig. 1. Location of sustainable ecological pond management in campus, Miaoli, Taiwan. Literature Review and Experts Consultation The assessment criteria for the campus’ eco-pond Establish Hierarchical Structure Set up a Fuzzy Positive Reciprocal Matrix Fig. 2 The operation steps of FAHP Campus, Miaoli, Taiwan. Group Integration Calculate the Fuzzy Weight Economy Management Management Planning Regular Maintenance Health and Sanitation Personnel Quality Environment Education Fig. 3. Hierarchy structure for assessment of FAHP campus, Malio, Taiwan 6 Defuzzifier Hierarchical Cascade and Ranking Fig. 2. The operation steps of FAHP in campus, Miaoli, Taiwan. ISBN: 978-960-6766-87-9 67 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 Table1 Results of expert group decision evaluation using FAHP in sustainable ecological pond management in campus, Miaoli, Taiwan. High Level (Theme) Ecosystem Condition (32.79%a1) Benchmark 1.Site and Layout 2.Natural Environment 3.Cultural Heritage 4.Traffic Conditions 5.Existing Biology 6. Groundwater System 7. Water in Pond 8. Infiltrated in Pond Bottom 9. Soil type and Environment 10. Benefit to Ecology 11. Plant in Water 12. Benefit to Biodiversity 13. Purify Water Quality by Microbiology 14. Safety Facility 15. Economy Management 16. Regular Maintenance 17. Health and Sanitation 18. Personnel Quality 19. Environmental Education Defuzzification 0.1974 0.2662 0.0986 0.0533 0.1117 0.2728 0.2760 0.1721 0.1157 0.1341 0.1502 0.0783 0.0737 0.2099 0.1089 0.2212 0.1234 0.1495 0.1871 Total weight 0.0647 0.0873 0.0323 0.0175 0.0366 0.0895 0.0561 0.0350 0.0235 0.0273 0.0305 0.0159 0.0150 0.0984 0.0511 0.1037 0.0578 0.0701 0.0877 Corrected Value 0.0204 0.0276 0.0102 0.0055 0.0116 0.0282 0.0207 0.0129 0.0087 0.0100 0.0112 0.0059 0.0055 0.0311 0.0161 0.0327 0.0183 0.0221 0.0277 Weighted R. 8 5 14 19 12 3 7 11 16 15 13 17 18 2 10 1 9 6 4 Soil and Water (20.33%a2) Management Planning (46.88%a3) a1, a2, a3 means “ecosystem condition” , “soil and water”, and “management planning” can be ranked and weighted collectively for 32.79%, 20.33% and 46.88 %, respectively. 0.03 0.025 0.02 0.015 0.01 0.005 0 Si te an N d at La ur yo al ut En vi ro nm Cu en ltu t ra lH Tr er ita af fic ge Co nd U iti Ex nd on ist er s in gr g ou Bi nd ol og W y at Ca er n Sy be ste In W m fil ate tra te ri d n in Po So Po nd il nd ty pe Bo an tto d m En vi ro Be nm ne en fit t to Ec ol og Pl y an Be ti ne n fit W to at er Bi od Pu iv rif er y sit W y at er Q ua lit Sa y Ec fe ty on Fa om cil y ity M an Re ag gu em lar en M t ain H ea te lth na nc an e d Sa ni Pe ta tio rs En on n vi ne ro lQ nm ua en lit tal y Ed uc at io n Fig. 4. Level of towards sustainable ecological pond management in campus, Miaoli, Taiwan. Left column shows relative weight of the 19 benchmarks. 7 ISBN: 978-960-6766-87-9 68 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 Non-linear Multi-objective Transportation Problem: A Fuzzy Goal Programming Approach H.R. MALEKI Shiraz University of Technology Department of Basic Sciences P. O. Box 71555-313 Shiraz IRAN maleki@sutech.ac.ir S. KHODAPARASTI Shiraz University of Technology Department of Basic Sciences P. O. Box 71555-313 Shiraz IRAN sa.khodaparasti@sutech.ac.ir Abstract: A non-linear multi-objective transportation problem (NMOTP) refers to a special class of non-linear multi-objective problems. In this paper we review goal programming (GP) and fuzzy programming (FP) as two approaches for solving (NMOTP). Meanwhile, by extending Mohamed’s idea about fuzzy goal programming (FGP), we propose a fuzzy goal programming approach to solve the non-linear multi-objective transportation problem. To this end, we use a special type of non-linear (hyperbolic) membership functions. Then, by proving a theorem, the relationship between (FP) and (FGP) will be stated. At the end, a numerical example is given to illustrate the efficiency of the proposed approach. Key–Words: Multi-objective programming problem, non-linear transportation problem, goal programming, fuzzy programming, non-linear membership function (hyperbolic). 1 Introduction In the real-world situations, the transportation problem usually involves non-linear, non-commensurable, multiple and conflicting objective functions. This kind of problem is called non-linear multi-objective transportation problem (NMOTP). Some authors apply a distance function to introduce the mathematical model of the non-linear multi-objective transportation problem. In this case the objective functions of the transportation problem become nonlinear [3]. After introducing the concept of fuzzy set theory by Zadeh [16] in 1965, Zimmermann [17] applied the fuzzy programming technique with some suitable membership functions to solve multi-objective linear programming problems. The results obtained by fuzzy linear programming lead to efficient solutions, too. Bit et al. [5] by using linear membership function, applied the fuzzy programming technique to solve multiobjective transportation problem. In 1999, Biswal and Verma [3] used fuzzy programming technique to find the optimal compromise solution of a nonlinear multiobjective transportation problem. Other approaches in this area have also introduced [4, 6]. The goal programming approach was first introduced by Charnes and Cooper [7] and developed goal programming introduced by Ignizio [10] was also widely used by other authors to solve multi-objective fuzzy programming problems [1, 8, 12, 14]. In 1997, Mohamed [13] presented the fuzzy goal programming approach to solve multi-objective linear programming problems. The main aim of this paper is extending Mohamed’s method to solve non-linear multi-objective transportation problems. To this end, we apply a special kind of membership function as hyperbolic. For the first time Leberling [11] used this type of membership function for the vector-maximum problem and showed that solutions obtained by hyperbolic membership function are equivalent to those obtained by linear one’s. The paper is organized as follows: In section 2, a mathematical model of non-linear multi-objective transportation problem is formulated. After, goal programming approach and fuzzy programming approach for solving non-linear multi-objective transportation problem are reviewed. In section 3, we propose a fuzzy goal programming approach to solve non-linear multi-objective transportation problems and a theorem about the relationship between (FP) and (FGP) is established. Finally, an example is presented to illustrate the efficiency of the proposed approach. 2 Mathematical Formulation A frequently encountered problem is the optimal location of centers of activities. This may involve the location of machines or departments in a factory, the ISBN: 978-960-6766-87-9 69 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 location of factories or warehouses from which goods can be shipped to retailers or consumers, or the location of emergency facilities ( i.e fire or police stations) in an urban area [2]. Consider the following simple case: Suppose that there n markets with known demands and locations. These demands are to be met from m warehouses of known capacities. The problem is to determine the location of the warehouses so that the total distance weighted by the shipment from the warehouses to the markets is minimized. More specially, let (xi , yi ) : unknown location of warehouse i for i = 1, 2, · · · , m, (aj , bj ) : known location of market j for j = 1, 2, · · · , n, ci : capacity of warehouse i for i = 1, 2, · · · , m, ej : known demand at market j for j = 1, 2, · · · , n, wij : units shipped from warehouse i to market area j for i = 1, 2, · · · , m; j = 1, 2, · · · , n, dij : distance from warehouse i to market area j for i = 1, 2, · · · , m; j = 1, 2, · · · , n. cr : penalty associated with transporting a unit of ij product from source i to destination j, with respect to r-th objective, for i = 1, 2, · · · , m; j = 1, 2, · · · , n. The mathematical model of (NMOTP) can be stated as follows: min : Zr = m n ¯ ¯ ¯ d+ = max(0, Zr − Zr ) = 1 {(Zr − Zr ) + |Zr − Zr |}, r 2 ¯ ¯ ¯ d− = max(0, Zr − Zr ) = 1 {(Zr − Zr ) + |Zr − Zr |}. r 2 for r = 1, 2, · · · , k. As minimization problem needs ¯ Zr ≤ Zr , therefore minimizing the distance between ¯ Zr and Zr leads to minimizing d+ . In this case if we r use the min-max form of goal programming, then the model (1) is converted to the following non-linear programming model: min : ψ, n s.t. j=1 m wij ≤ ci , wij ≤ ej , i = 1, 2, · · · , m, j = 1, 2, · · · , n, r = 1, 2, · · · , k, r = 1, 2, · · · , k, (2) i=1 ¯ Zr + d− − d+ = Zr , r r ψ ≥ d+ , r d+ , d− ≥ 0, d+ .d− = 0, r = 1, 2, · · · , k, r r r r wij , xi , yi ≥ 0 for all i, j, i=1 j=1 n cr wij dij , r = 1, 2, · · · , k, ij i = 1, 2, · · · , m, j = 1, 2, · · · n, for all i, j, (1) where, dij = [(xi − aj )2 + (yi − bj )2 ]1/2 , and the equilibrium condition m ci ≥ n ej is satisfied. i=1 j=1 2.2 Fuzzy Programming s.t. j=1 m i=1 wij ≤ ci , wij ≤ ej , wij , xi , yi ≥ 0 where, dij = [(xi − aj )2 + (yi − bj )2 ]1/2 , and the equilibrium condition m ci ≥ n ej is satisfied. i=1 j=1 2.1 Goal Programming Since the pioneering work on the concept of goal programming by Charnes and Cooper [7], several research results by different authors were reported [10, 12, 14]. The main idea behind the goal programming is to minimize distance between Z = (Z1 , Z2 , · · · , Zk ) and aspiration (goal) levels vector, say ¯ ¯ ¯ ¯ Z = (Z1 , Z2 , · · · , Zk ), that usually is determined by the decision maker. To express the distance between ¯ Z and Z, positive and negative deviational variables are introduced respectively as follows: Let Ur and Lr be the upper and lower bounds for the r-th objective, where Lr is the aspired level of achievement for the r-th objective function (best value) and Ur be the highest acceptable level of achievement for the r-th objective function (worst value). Algorithm: Step 1. Solve the multi-objective non-linear transportation problem as a single objective non-linear programming problem, taking each time only one objective as objective function and ignoring all others. Step 2. Compute the value of each objective function at each solution derived in Step 1. Step 3. By using results of Step 2, the best (Lr ) and the worst (Ur ) values for any objective functions are determined. If we use these results as aspiration levels, then the initial fuzzy model can be formulated as: Find wij , xi , yi which satisfy: Zr < Lr , ∼ r = 1, 2, · · · , k, ISBN: 978-960-6766-87-9 70 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 n j=1 m i=1 wij ≤ ci , i = 1, 2, · · · , m, wij ≤ ej , j = 1, 2, · · · n, for all i, j, (3) function is 1, so by introducing the deviational variables d− , d+ ≥ 0 corresponding to the r-th hyperbolic memberr r ship function defined in (4), the flexible membership goal with aspired level 1 can be described as: 1 1 e{Jr −Zr }αr − e−{Jr −Zr }αr + + d− − d+ = 1, r r 2 2 e{Jr −Zr }αr + e−{Jr −Zr }αr where d− · d+ = 0. r r Also any over-deviation from the fuzzy goal “1” indicates the full achievement of the membership value. Therefore, to achieve the aspired level of the fuzzy goal, it is enough to minimize its negative deviational variable from “1”. Applying the Min-Max form of goal programming to the model of the multi-objective non-linear transportation (1), one can obtain the following fuzzy goal programming model formulation: min : φ, 1 1 e{Jr −Zr }αr − e−{Jr −Zr }αr + + d− − d+ = 1, s.t. r r 2 2 e{Jr −Zr }αr + e−{Jr −Zr }αr r = 1, 2, · · · , k, φ≥ n j=1 m wij , xi , yi ≥ 0 where, dij = [(xi − aj )2 + (yi − bj )2 ]1/2 , and the equilibrium condition m ci ≥ n ej , is satisi=1 j=1 fied. Note that the symbol “< ” is the fuzzification of ∼ “≤” and the notation Zr < Lr stands for “Zr is sub∼ stantially smaller than Lr ” and which is represented by an upper semi-continuous non-increasing membership function μr (Zr ) [11]. To solve (3), a hyperbolic membership function μH (Zr ) corresponding to r-th objective function is der fined in [11] as: μH (Zr ) = r ⎧ ⎨ 1 1 1 e{ Jr −Zr }αr −e−{Jr −Zr }αr 2 + 2 e{Jr −Zr }αr +e−{Jr −Zr }αr ⎩ 0 6 where, αr = Ur −Lr , Jr = Ur +Lr . 2 if Zr ≤ Lr , if Lr < Zr < Ur , (4) if Zr ≥ Ur , Now, for model (3), by using the “ Zadeh min ” operator with the membership function (4), we obtain an equivalent crisp non-linear programming model as follows: max : λ, 1 1 e{Jr −Zr }αr − e−{Jr −Zr }αr , s.t. λ ≤ + 2 2 e{Jr −Zr }αr + e−{Jr −Zr }αr r = 1, 2, · · · , k, n d− , r (6) r = 1, 2, · · · , k, i = 1, 2, · · · , m, j = 1, 2, · · · , n, wij ≤ ci , wij ≤ ej , i=1 φ ≤ 1, (5) d+ , d− ≥ 0, d+ .d− = 0, r r r r φ ≥ 0, wij , xi , yi ≥ 0 r = 1, 2, · · · , k, for all i, j, wij ≤ ci , j=1 m i = 1, 2, · · · , m, j = 1, 2, · · · , n, where, dij = [(xi − aj )2 + (yi − bj )2 ]1/2 , and the m n equilibrium condition i=1 ci ≥ j=1 ej , is satisfied. The following theorem establishes the equivalence between models (5) and (6). Theorem 1 The crisp non-linear programming model (5) obtained by using “Zadeh min” operator with membership function (4) for the multi-objective non-linear transportation problem (1) is equivalent to the fuzzy goal programming model (6). Proof. At first we rewrite the model (5) as: min : 1 − λ, s.t. 1 1 e{Jr −Zr }αr − e−{Jr −Zr }αr }, 1−λ≥1−{ + 2 2 e{Jr −Zr }αr + e−{Jr −Zr }αr n j=1 wij ≤ ej , i=1 λ ≤ 1, λ ≥ 0, wij , xi , yi ≥ 0 for all i, j, where dij = [(xi − aj )2 + (yi − bj )2 ]1/2 , and the equilibm n rium condition i=1 ci ≥ j=1 ej , is satisfied. 3 Fuzzy Goal Programming Introducing fuzzy goal programming approach [9, 15] for solving multi-objective linear programming problem, Mohamed [13] has used linear membership function. In this section we adopt his approach to extend it for solving a multi-objective non-linear transportation problem defined in (1). Since that the highest degree of any membership r = 1, 2, · · · , k, wij ≤ ci , (7) i = 1, 2, · · · , m, ISBN: 978-960-6766-87-9 71 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 m wij ≤ ej , i=1 j = 1, 2, · · · , n, φ ≤ 1, φ ≥ 0 wij , xi , yi ≥ 0 for all i, j. λ ≤ 1, λ ≥ 0, wij , xi , yi ≥ 0 for all i, j. where, dij = [(xi − aj )2 + (yi − bj )2 ]1/2 , by satisfying the equilibrium condition i=1 ci ≥ So the proof is totally perfect. m n j=1 ej . where, dij = [(xi − aj )2 + (yi − bj )2 ]1/2 , and the m n equilibrium condition i=1 ci ≥ j=1 ej , is satisfied. By setting φ = 1 − λ, we have: min : φ, s.t. 1 1 e{Jr −Zr }αr − e−{Jr −Zr }αr φ≥1−{ + }, 2 2 e{Jr −Zr }αr + e−{Jr −Zr }αr n Example: To illustrate the efficiency of the proposed method we consider the following problem which is used by Biswal et al. [3]. min : Z1 = 10w11 [(x1 − 6)2 + (0 − 5)2 ]1/2 + 10w12 [(x1 − 10)2 + (0 − 5)2 ]1/2 + 10w13 [(x1 − 9)2 + (0 − 11)2 ]1/2 + 10w14 [(x1 − 14)2 + (0 − 8)2 ]1/2 + 20w21 [(0 − 6)2 + (y2 − 5)2 ]1/2 + 20w22 [(0 − 10)2 + (y2 − 5)2 ]1/2 + 20w23 [(0 − 9)2 + (y2 − 11)2 ]1/2 + 20w24 [(0 − 14)2 + (y2 − 8)2 ]1/2 , min : Z2 = 18w11 [(x1 − 6)2 + (0 − 5)2 ]1/2 + 18w12 [(x1 − 10)2 + (0 − 5)2 ]1/2 + 18w13 [(x1 − 9)2 + (0 − 11)2 ]1/2 + 18w14 [(x1 − 14)2 + (0 − 8)2 ]1/2 + 6w21 [(0 − 6)2 + (y2 − 5)2 ]1/2 + 6w22 [(0 − 10)2 + (y2 − 5)2 ]1/2 + 6w23 [(0 − 9)2 + (y2 − 11)2 ]1/2 + 6w24 [(0 − 14)2 + (y2 − 8)2 ]1/2 , 4 j=1 r = 1, 2, · · · , k, wij ≤ ci , i = 1, 2, · · · , m, (8) j=1 m wij ≤ ej , i=1 j = 1, 2, · · · , n, φ ≤ 1, φ ≥ 0, wij , xi , yi ≥ 0 for all i, j. w1j ≤ 50, wi1 = 20, wi3 = 60, 4 j=1 2 i=1 2 i=1 w2j ≤ 100, 2 i=1 2 i=1 wi2 = 40, wi4 = 30, where, dij = [(xi − aj )2 + (yi − bj )2 ]1/2 , and the m n equilibrium condition i=1 ci ≥ j=1 ej , is satisfied. Let’s define the negative and positive deviational variables d− and d+ as: r r {Jr −Zr }αr −{Jr −Zr }αr d− = max{0, 1 − ( 1 + 1 e{Jr −Zr }αr −e−{Jr −Zr }αr )}, r 2 2e +e d+ = max{0, ( 1 + r 2 1 e{Jr −Zr }αr −e−{Jr −Zr }αr 2 e{Jr −Zr }αr +e−{Jr −Zr }αr ) − Then from model (8), we have φ ≥ d− ,where r 1 1 e{Jr −Zr }αr −e−{Jr −Zr }αr − + 2 + 2 e{Jr −Zr }αr +e−{Jr −Zr }αr + dr − dr = 1. wij , xi , yi ≥ 0, i = 1, 2 ; j = 1, 2, 3, 4. 1}. It is assumed that the first source is at (x1 , y1 ) and the second source is at (x2 , y2 ), where the cost matrices are given as: C1 = 10 20 10 10 20 20 10 20 , C2 = 18 6 18 6 18 6 18 6 . Finally the model (8) can be rewritten as: min : φ, 1 1 e{Jr −Zr }αr − e−{Jr −Zr }αr s.t. + + d− − d+ = 1, r r 2 2 e{Jr −Zr }αr + e−{Jr −Zr }αr d− · d+ = 0, r r φ≥ n j=1 m The co-ordinates of the four markets with reference to X and Y axis are given as where the distance is measured in kilometers. (a1 , b1 ) = (6, 5) , (a2 , b2 ) = (10, 5), (a3 , b3 ) = (9, 11) , (a4 , b4 ) = (14, 8). Using the fuzzy goal programming method with hyperbolic membership function, the problem is solved and the optimal solution is obtained as: ∗ ∗ w11 = 0.000000, w12 = 37.564875, ∗ ∗ w13 = 0.000000, w14 = 12.435125, ∗ ∗ w21 = 20.000000, w22 = 2.435125, ∗ ∗ w23 = 20.000000, w24 = 17.564875, r = 1, 2, · · · , k, r = 1, 2, · · · , k, (9) d− , r wij ≤ ci , i = 1, 2, · · · , m, (x1 , y1 ) = (10.642053, 0.000000), wij ≤ ej , j = 1, 2, · · · , n, (x2 , y2 ) = (0.000000, 8.972148), ∗ ∗ Z1 = 22375.63357, Z2 = 11171.52638. i=1 ISBN: 978-960-6766-87-9 72 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 4 Conclusion In this paper a fuzzy goal programming approach is developed to solve an special mathematical model of non-linear multi-objective transportation problem. In goal programming models absolute deviations are obtained in the optimal solution. Another benefit of the models based on goal programming is the minimum changes required for sensitively analysis. Fuzzy goal programming is used when a decision maker is unable to specify precise aspiration levels and hence changes of admissible violations may frequently occur. The other methods can be affected by these changes more than goal programming methods. Acknowledgements: This paper is supported in part by Fuzzy Systems and Applications Center of Excellence, Shahid bahonar University of Kerman, Kerman, I.R. of Iran. [13] R.H. Mohamed, The relationship between goal programming and fuzzy programming, Fuzzy Sets and Systems. 89, 1997, pp. 215–222. [14] R. Narasimhan, Goal programming in a fuzzy environment, Decision Sci. 11, 1980, pp. 325–336. [15] R.N. Tiwari and S. Dharmar and J.R. Rao, Fuzzy goal programming-an additive model, Fuzzy Sets and Systems. 24, 1987, pp. 27–34. [16] L.A. Zadeh, Fuzzy sets, Inform. and Control. 8, 1965, pp. 338–353. [17] H.J. Zimmermann, Fuzzy programming and linear programming with several objective functions, Fuzzy Sets and Systems. 1, 1978, pp. 45–55. References: [1] R.S. Aenaida and N.W. Kwak, A linear goal programming for transshipment problems with flexible supply and demand constraints, Operation Research Society. 45(2), 1994, pp. 215–224. [2] M.S. Bazaraa and C.M. Shetty, Non-linear programming, John Wiley and Sons – New York – 1979. [3] M.P. Biswal and R. Verma, Fuzzy Programming Technique to Solve A Non-Linear Transportation Problem, Fuzzy Mathematics. 7, 1999, pp. 723–730 [4] A.K. Bit, Multi-objective Geometric Programming Problems:Fuzzy Programming with Hyperbolic Membership Function, Fuzzy Mathematics. 6,1998, pp. 27–32 [5] A.K. Bit and M.P. Biswal and S.S Alam, Fuzzy programming approach to multi-criteria decision making transportation problem, Fuzzy Sets and Systems. 50, 1992, pp. 135–141. [6] S. Chanas and D. Kuchta, A concept of the optimal solution of the transportation problem with fuzzy cost coefficients, Fuzzy Sets and Systems. 82, 1996, pp. 299–305. [7] A. Charnes and W.W. Cooper, Management Models of Industrial Applications of Linear Programming (Appendix B),Vol. I, Wiley–New York–1961. [8] E.L. Hannan, Linear programming with multiple fuzzy goals, Fuzzy Sets and Systems. 6, 1981, pp. 235–248. [9] E.L. Hannan, On fuzzy goal programming, Decision Sci. 12, 1981, pp. 522-531. [10] J.P. Ignizio, Goal Programming and Extensions, Lexington D.C. Health, MA, 1976. [11] H. Leberling, On finding compromise solutions in multi-critical problems using the fuzzy min–operator, Fuzzy Sets and Systems. 6, 1981, pp. 105–118. [12] S.M. Lee and L.J. Moore, Optimizing transportation problems with multiple objectives, AIEE Transactions. 5, 1973, pp. 333–338 ISBN: 978-960-6766-87-9 73 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 Drivers’ stated preferences towards road pricing: the case of Athens Christina Milioti Department of Transportation Planning and Engineering National Technical University of Athens 5 Iroon Polytechniou Street, 15773, Zografou, Athens GREECE christinamilioti@hotmail.com Ioanna Spyropoulou Department of Transportation Planning and Engineering National Technical University of Athens 5 Iroon Polytechniou Street, 15773, Zografou, Athens GREECE iospyropop@central.ntua.gr Matthew Karlaftis Department of Transportation Planning and Engineering National Technical University of Athens 5 Iroon Polytechniou Street, 15773, Zografou, Athens GREECE mgk@central.ntua.gr Abstract: - Traffic congestion is a phenomenon which threatens the sustainable development of urban areas, and measures to tackle the problem are being sought. A quite effective way targeting traffic congestion is road pricing. The principle of road pricing is to charge for the “burden” that road users produce mainly on traffic and subsequently on environmental conditions. Within this study a road pricing scheme has been designed for the city of Athens, to investigate driver attitudes towards the scheme, through a questionnaire survey which was conducted to travellers in the city of Athens. Participants of the survey were asked to answer several questions to elicit their preferences in respect to road pricing, their acceptability and their willingness-to-pay for the implementation of such a scheme. Results indicate that the majority of travellers is aware of the advantages of a road pricing scheme and is not against (about 10% in favor and about 53% conditionally agree) the design of such a measure to reduce urban congestion and that there are specific parameters that influence scheme acceptability. Key-Words: Congestion,Congestion charging, Acceptability, Stated preference 1 Introduction Congestion is a common phenomenon in urban centres which threatens the sustainability of the cities and a key political issue in many European countries [1]. Congestion is a consequence of the increase in the demand for travel and of the increase in car ownership, and effective measures for managing the demand and urban congestions are sought. The impacts of congestion are not restricted to traffic conditions (increased travel time, queue lengths, delays) but also include traffic accidents, traffic noise, environmental pollution (air pollutants), parking problems, impact on health etc [2]. Hence, the European Commission has issued the Green Bible to raise awareness and organise specific policy and actions towards a new culture for urban mobility mainly targeting at reducing urban congestion, improving environmental conditions, increasing public transport patronage, walking and cycling [3]. Road pricing has emerged as an effective measure for tackling urban congestion (and the consequent ISBN: 978-960-6766-87-9 74 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 environmental pollution) and has been implemented in several cities including London, Rome, Milan, Stockholm, Bergen, Oslo and Trodheim [4]. The effectiveness of a road pricing scheme depends largely on public acceptability which denotes an evaluation of the expected impacts of the scheme by the users (residents, car drivers, public transport passengers, businesses etc.) [5]. The purpose of this study is to investigate drivers’ preferences towards congestion pricing under a hypothetical such scheme for the city of Athens. variables/characteristics – namely, toll price, travel time reduction and travel time deviation as illustrated in Figure1. Nine different such scenarios were presented and respondents asked to note the route they would choose in each scenario considering a typical trip to the centre of Athens by car on a weekday between 7.00am and 9.00pm of 45 minute duration. *** Please Insert Figure 1 here*** Finally, within the fourth section of the questionnaire data on the demographic and socioeconomic characteristics of the drivers including gender, age, household income, occupation and level of education was obtained. 2 Methodology 2.1 The questionnaire The main tool of this study was a stated preference questionnaire, which was organised in four sections. The first section involved information related to elements of respondents travel behaviour in respect to their trip characteristics. In particular, the frequency, purpose and duration of the most commonly followed route to the center of Athens were obtained. In addition, respondents also provided information on their origin and destination locations. The necessity of obtaining such information is based on the assumption that traveller acceptability and willingness-to-pay for a congestion charging scheme is dependent on trip characteristics. The objective of the second section of the questionnaire was twofold. First, as respondents had not experienced the operation of a congestion charging scheme before, this section aimed at increasing their awareness on such as scheme and on its potential impact both positive and negative. Second, as driver perceptions on traffic congestion and road pricing may also affect their acceptability and willingness-to-pay the aim of this section was to induce such dependencies. Hence, this section contained questions related to the impact that congestion has on respondents’ activities, possible measures to improve traffic conditions, advantages and disadvantages of a congestion pricing scheme, potential use of the scheme revenues. The last question of this section asked whether respondents would be in favor of or against to the implementation of a congestion pricing scheme. The stated preference experiment comprised the third part of the questionnaire, and its objective was to generate data for a discrete choice analysis to investigate drivers’ willingness-to-pay. In this section, respondents were presented with two different routes defined by three different 2.2 The stated preference experiment The first property of a stated preference experiment to be determined involves the variables to be considered that describe the presented choices, and in the examined case the route characteristics. The variables should be chosen under the assumption that they affect driver choice. Next the number and value of the variable attribute levels needs to be defined. The attribute levels need to be realistic and cover a substantial part of the whole spectrum of possible values. Increasing the number of variables and attribute levels results to a more detailed experiment but at the same time to a complex and time consuming experiment. In the presented study, it was decided to employ three variables each of which was described by three attribute levels. A pilot survey was conducted to identify the most crucial parameters that determined driver willingness-to-pay. The two primary variables to determine drivers’ preferences, which were an obvious choice, were toll price and travel time reduction. Two other variables that were considered were use of revenues the three different attribute levels of which being public transport improvement, social causes and tax reduction, and travel time deviation. The analysis of the pilot questionnaire data indicated that the use of revenues for public transport improvement was the obvious choice for most respondents and therefore this variable did not provide any additional effect on drivers’ preferences and was excluded from the discrete choice experiment. The three variables finally selected for the experiment were: toll price, travel time reduction and travel time deviation. Each of these variables had three different attribute levels which are illustrated in Table1. *** Please Insert Table 1 here*** Hence, there are 33 =27 combinations of all possible values and 351 possible pairs of these combinations that result for use in a binary choice ISBN: 978-960-6766-87-9 75 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 experiment Therefore the following fractional factorial design experiment was derived. The 27 combinations were partitioned in three blocks A, B, C, each of which consisted of nine profiles. A constraint was applied for the allocation of the different profiles into the three blocks, which was to ensure a balanced design in each block. A design is balanced when for each variable each attribute level appears in all blocks the same numbers of times, and is usually considered to be a prerequisite for efficient designs. The binary choice experiment was achieved by matching the profiles of two blocks (A-B, A-C and B-C) under two constraints. Obtaining an orthogonal design and minimising the number of dominant choices. Orthogonality ensures that the variables used are varied independently from one another [6], and comprises a characteristic of an efficient design. Still, orthogonality properties may not be retained in the estimation process especially when there are non-competed choices within the stated preference section of a questionnaire [7]. Each one of the three matching A-B, B-C, A-C obtained this way appeared as a set of nine binary choices in the stated preference section of the questionnaire. The pairing of the profiles was achieved by modeling the problem as a 0-1 integer programming problem, which in fact is an assignment problem having as side constrains the conditions for orthogonality and, as objective function to be minimised, the sum of the assignment variables corresponding to the obvious choices. Only a very limited, 3 in the worst case scenario, dominant choices appeared in the matching. Thus the information loss was not significant compared to the advantage of the elimination of cross-effects that orthogonality claims. The three different choice sets A-B, B-C and A-C appear in Table 2. *** Please Insert Table 2 here*** 3 Field Survey The questionnaire was completed by direct interviews made in person. The only quota applied to the sample was that the respondents were passenger-car drivers entering the center of Athens. The field survey was conducted on working days, at different working hours of the day in an attempt to capture different driver and trip characteristics. Most interviews were carried out in the center of Athens mainly in offices and in organised parking stations, so that the respondents could devote all the required time for answering the questions. During the interview and completion of the questionnaire detailed explanations were provided to make sure that respondents had comprehended all the issues concerning congestion pricing. The interviewers gave detailed explanations to all respondents’ questions, so as to prevent any ambiguity on the provided answers. 262 questionnaires were completed and after a detailed study of the answers given, 22 of them were found to contain either mistakes or inconsistent answers and were not included in the analysis. Hence, the final sample consisted of 240 questionnaires; and Table 3 summarises the socioeconomic and demographic characteristics of the sample. *** Please Insert Table 3 here*** Driver trip characteristics were also analysed and results are presented in Table 4. *** Please Insert Table 4 here*** The majority of drivers driving to the centre of Athens makes work related trips (68%), 75.4% of which are made on a daily basis comprising commuting trips. This indicates that the majority of trips made to the centre of Athens by car involve non-elastic trips, i.e. trips the characteristics of which including frequency, destination and time of arrival at destination cannot be modified. 25.4% of trips involve leisure or shopping purposes and 60% of which are made rather occasionally (<2 times in a week). Hence, drivers making leisure/shopping trips might be able to adapt their trip characteristics under new conditions, e.g. implementation of a road pricing scheme. A small proportion of drivers made trips related to studying locations (e.g. university, college etc). The reason for this category being introduced rather than being incorporated within the work related trips is that the characteristics of such trips are different to work related trips in that their frequency and time of arrival at destination point can be adaptable to new conditions. More than half of the respondent trips were commuting trips (54.6%), and the rest were almost evenly distributed between frequent and occasional trips. 4 Driver attitudes 4.1 Driver preferences on road pricing Statistical analysis of data extracted from the second section of the questionnaire elicited driver preferences in relation to the measure of road pricing. What needs to be considered together with the identified trends is that the survey participants have not experienced the implementation and effects of such a measure. Respondents experience congested traffic conditions, as the road network in ISBN: 978-960-6766-87-9 76 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 Athens suffers from heavy congestion during the peak hours. Increased travel time was considered to be the most important effect of traffic congestion by the participants being followed by environmental pollution and deterioration of psychological calm. The advantages and disadvantages of an urban congestion scheme as perceived by the road users, provide an indication of user acceptability factors. Tables 5 and 6 illustrate drivers perceptions on the advantages and disadvantages of a congestion pricing scheme (participants were asked to rank their answers in order of preference). *** Please Insert Table 5 here*** The perceived advantages of a road pricing scheme mainly involved the mitigation of traffic congestion (about 35% selected it as the most or second most important advantage) and the improvement of environmental conditions (about 30% selected it as the most or second most important advantage). The reduction of traffic congestion and, in fewer cases, the reduction of environmental pollution comprises the primary objective of the majority of the implemented urban congestion schemes [1]. This comes in accordance with driver perceptions. In certain cases, in Norwegian cities (Oslo, Trodheim and Bergen), the primary aim of such a scheme was the collection of revenue to be invested first on improving road infrastructure and secondly on public transport services [8]. Revenue utilisation scored rather low in drivers’ perceptions on the advantages of a road pricing scheme as about 10% of the drivers selected it as the most or second most important advantage. In addition, about 10% of the drivers did not consider revenue utilisation as an advantage. About 16% considered that the most or second most important advantage of a congestion pricing scheme is the quality of life improvement which is a qualitative effect that may include the aforementioned ones. A small proportion of drivers considered attributing cost to those who create it as a significant advantage. Last, 2.5% of the drivers are of the view that a congestion pricing scheme does not present any advantages. *** Please Insert Table 6 here*** As anticipated, the vast majority of drivers considered the user cost – as expressed through the toll cost – of a road pricing scheme as the most crucial disadvantage (43% ranked it as 1st or 2nd choice). The value of the toll determines several elements of the scheme including its acceptability, the resulting reduction of car-use and the amount of collected revenue. Second most important disadvantage was perceived to be the loss of privacy that users perceive to be a result of them being monitored when entering or exiting the scheme area, being followed by the scheme implementation costs. There is a small, yet existent, proportion of participants (about 6%) who were of the view that the implementation of a road pricing scheme does not have any adverse effects. Findings in similar studies indicated that the use of the collected revenues of a congestion pricing scheme is a crucial factor which affects scheme acceptability [9]. Table 7 illustrates respondent preferences on the use of revenues of a congestion pricing scheme. *** Please Insert Table 7 here*** The great majority of respondents prefer the use of revenues in sectors that will result in an improvement of travel conditions, as about 53% answered that the revenues should be used towards public transport and road network improvements (as 1st or 2nd choice). The most popular sector for revenue use proved to be public transport improvement (about 34%). This can be explained as the desire of drivers to be provided with alternative transport modes if car-usage is charged, which also involves a well-defined, effective and retributive use of revenues. More general uses of revenues such as tax reduction and social causes show low priorities in the respondents’ preferences. 4.2 Driver acceptability on road pricing Driver acceptability of a road pricing scheme is a prerequisite for its successful implementation. It has been observed however, that the acceptability increases following scheme implementation. For the city of Athens, 10.8% of the respondents were in favor of congestion pricing, 32.1% were against it while 53.3% conditionally agreed. The latter category expressed the approval of road pricing with the prerequisite of additional actions to support the measures; for example improving the services of alternatives to car-use, or using the revenues to improve public transport. To interpret the factors that affect acceptability statistical analysis was performed. In particular a binary logistic model was developed having acceptance as a dependent variable and variables related to driver characteristics, trip characteristics and driver perceptions as independent ones. The results indicated that driver acceptance depended on the factors demonstrated in Table 8. *** Please Insert Table 8 here*** Trip frequency and trip duration are positively related to acceptability; hence, increase of these two variables increases acceptability. This is justified as increased trip duration and trip frequency results in the impact of improved travel conditions, which is ISBN: 978-960-6766-87-9 77 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 anticipated as a consequence of congestion pricing, being more evident to the driver. Trip purpose however did not affect driver acceptability significantly, although in principle the consequences of congestion should affect more people travelling to work than people travelling for leisure or shopping. Driver characteristics were also not found to affect driver acceptability. Acceptability of the system is influenced by use of revenues [10], [11], [12]. In several cases it has been established that acceptability is positively related to the use of revenues towards improving public transport services. This is because those who accept the pricing scheme consider the improvement of travel conditions of primary importance and hence a concrete use of revenues would more efficient. This comes in accordance with the findings from surveys in London, which demonstrate that acceptability of urban pricing increases from 43% to 63%, if the revenues are used for transport and environmental improvements [13]. road pricing acceptability shows that those who agree with the pricing scheme (even conditionally agree) are willing to pay a higher toll. Drivers who consider the using the scheme revenues to towards tax reduction usually prefer the lower toll scenario. Last, demographic characteristics such as respondent gender, age, marital status, education and income did not appear to influence driver choice regarding their willingness–to-pay. It should be noted that while filling-in the questionnaires respondents did not feel comfortable with declaring their household income, or were not fully aware of it. 5 Conclusion Congestion is considered as a serious problem affecting the sustainability of urban centres and urban road pricing is considered as an effective way for traffic management. Although the toll cost is considered a serious disadvantage of the system in view of the high acceptability, one can infer that the trade off between cost and benefits arising from a pricing scheme turns in favor of the benefits. Under this light this study investigated traveller preferences for a road pricing scheme for the city of Athens, with the use of a questionnaire survey. Participants indicated that they considered road pricing to be an effective means towards targeting both traffic congestion and environmental pollution. Only a small participant number (2,5%) noted that road pricing does not produce any advantages. In terms of disadvantages, user cost was considered to be the most crucial one, with loss of privacy and infrastructure cost being also considered as disadvantages. Concerning the use of the revenues, the great majority of the respondents prefer a specially purpose use aiming at improving public transport and environmental conditions and not a general purpose use such a tax reduction. In general, acceptability of the road pricing scheme can be considered to be quite high as approximately 60% of the participants agreed or conditionally agree with road pricing; and acceptability usually increases following scheme implementation. Acceptability on the scheme also depended on the trip patterns and in particular on trip duration and trip frequency, but was not found to be influenced by the aim of the trip. Demographic characteristics were also not found to affect acceptability. Travel time reduction resulting from the implementation of the scheme was found to be the most important factor influencing driver’s choice of a higher toll instead of a lower one. Travel time 4.3 Driver acceptability on road pricing This analysis was performed using as input data the responses in the stated preference experiment, and investigate drivers’ willingness-to-pay between a high and a low toll, corresponding to the two alternative routes presented to them in each examined scenario, in relation to specific route characteristics. To interpret drivers’ decision to accept a high or low toll level as a function of the travel conditions, a binary logistic regression model was developed having as a dependant variable the choice between a high or a low toll level and independent variables travel time reduction, travel time deviation, and other driver characteristics. Table 9 illustrates the developed model. *** Please Insert Table 9 here*** The resulting model produced an average fit (R2=0,624) and the correct prediction percentages were 86,8% (71,15% for high and 86,8% for toll selection). The results of the model indicate that the most important factors related to the driver’s willingness to pay decision are travel time reduction and travel time deviation. As it was expected value of time is a crucial factor which affects driver willingness to pay a high or low tariff level. A positive correlation between high toll level and driver perception that traffic comprises the main advantage of a pricing scheme was found, indicating that drivers who consider road pricing as a solution towards congestion are prepared to pay a high toll. Also a positive correlation between high toll and ISBN: 978-960-6766-87-9 78 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 deviation is somewhat less important, but still an important factor towards willingness-to-pay a higher instead of a lower toll. Last, drivers who considered as the main advantage of a road pricing scheme being the reduction of traffic congestion were more willing to pay a high toll. Demographic characteristics did not seem to affect willingness-topay. The results of this study highlighted several of the aspects that influence the acceptability of Athenians towards a road pricing scheme and References: [1] Jones, P., and Hervik, A., Restraining car traffic in European cities: an emerging role for road pricing, Transportation Research Part A, Vol.26, No.2, 1992, pp. 133-145. [2] E.C.M.T., The European conference of Ministers of Transport, Transport Policy and the environment, Paris, 1990. [3] Commission of the European Communities, Green Paper: Towards a new culture for urban mobility, Brussels, 2007. [4] Langmyhr, T., Understanding innovation: the case of road pricing, Transport Reviews, Vol.19, No.3, 1999, pp. 255-271. [5] Schuitema, G., and Steg, L., The role of revenue use in the acceptability of transport policies, Transportation Research Part F, Vol.11, No.3, 2008, pp. 221-231. [6] Hensher, D.A., Stated preference analysis of travel choices: The state of practice, Transportation, Vol.21, No.2, 1994, pp. 107-133. [7] Rose, J.M., and Bliemer, M.C.J., The design of stated choice experiments: The state of practice and future challenges, working paper, University of Sydney, 2004. [8] PATS, Empirical studies on price acceptability, Deliverable D.3, 2000. [9] Ieromonachou, P., Potter, S., and Warren J.P., A Strategic Niche analysis of urban road pricing in the UK and Norway, European Journal of Transport and Infrastructure Research, Vol.7, No.1, 2006, pp. 15-38. [10] Schade, J., and Schlag, B., Public acceptability of traffic demand management in Europe, Traffic Engineering and Control, Vol.41, 2000, pp. 314318. [11] Schade, J., and Schlag, B., Acceptability of urban transport pricing strategies, Transportation Research Part F, Vol.6, No.1, 2003, pp. 45-61. [12] Small, K.A., Using the revenues from congestion pricing, Transportation, Vol.19, No.4, 1992, pp. 359-381. [13] Jones, P.M., Gaining public sypport for road pricing through a package approach, Traffic Engineering and Control, Vol.32, 1991, pp. 194196. ISBN: 978-960-6766-87-9 79 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 Route option A Toll price (euros) 1 Route option B Toll price (euros) 2 Travel time reduction (%) 10 Travel time deviation (%) 10 Travel time reduction (%) 40 Travel time deviation (%) 30 Figure 1. Choice situation presented to the respondent Table 1. Variable attribute levels time Travel time Travel Toll price reduction deviation 1€ 10% 0% 2€ 20% 10% 4€ 40% 30% Table2. Choice sets of the state preference experiment Choice set A-B Binary choices Block Α Toll Time price reduction 10% 20% 40% 10% 10% 40% 20% 40% 20% Time deviation 30% 30% 10% 10% 0% 0% 0% 10% 30% Block Β Toll price 4€ 1€ 2€ 4€ 2€ 1€ 1€ 4€ 2€ Block C Toll price 4€ 1€ 2€ 4€ 1€ 2€ 2€ 1€ 4€ Block C Toll price Time reductio n 10% 10% 10% 20% 40% 40% 40% 20% 20% Time deviation 10% 0% 30% 10% 10% 30% 0% 0% 30% 1 4€ 2 4€ 3 4€ 4 2€ 5 2€ 6 2€ 7 1€ 8 1€ 9 1€ Choice set B-C Binary choices Block B Toll Time price reduction 10% 20% 20% 10% 20% 40% 10% 40% 40% Time deviation 10% 10% 0% 30% 30% 10% 0% 30% 0% 1 4€ 2 4€ 3 4€ 4 2€ 5 2€ 6 2€ 7 1€ 8 1€ 9 1€ Choice set A-C Binary choices Time reductio n 40% 10% 20% 10% 10% 20% 40% 20% 40% Time deviation 30% 10% 10% 0% 30% 0% 30% 10% 0% Block Α Toll Time price reduction Time deviation Time reductio Time deviation ISBN: 978-960-6766-87-9 80 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 1 2 3 4 5 6 7 8 9 4€ 4€ 4€ 2€ 2€ 2€ 1€ 1€ 1€ 10% 20% 40% 10% 10% 40% 20% 40% 20% 30% 30% 10% 10% 0% 0% 0% 10% 30% 4€ 4€ 1€ 2€ 1€ 1€ 4€ 2€ 2€ n 10% 40% 10% 40% 10% 20% 40% 20% 20% 0% 30% 10% 30% 30% 10% 0% 0% 10% Table 3. Sample characteristics Gender No. Age No. Employment Male 132 18 - 44 24 Female 108 25 - 91 34 35 - 50 45 >45 55 Employee Student Selfemployee Unemployed No. 162 25 45 8 Education No. 6 Household Income <800€ 800-2000€ No. 27 104 69 40 High-school University 57 177 20004000€ >4000€ Table 4. Characteristics of weekday trips to the centre of Athens Leisure/ Work Studies Total Shopping Every day 2-4 times a week <2 times a week Total 51.3% 12.1% 4.6% 68.0% 2.5% 7.5% 15.4% 25.4% 0.8% 5.0% 0.8% 6.6% 54.6% 24.6% 20.8% 100.0% 4th-5th option 25,83% 13,75% 57,50% 11,25% 61,67% not selected 5,83% 12,08% 10,83% not selected 10,00% 5,42% 9,58% 4,58% 14,17% - Table 5. Advantages of road pricing scheme 1st option Improved quality of life Reduction of environmental pollution Revenue utilisation Congestion improvement Paying toll for creating cost No advantage 15,00% 22,50% 8,75% 45,42% 5,83% 2,50% 2nd option 18,75% 37,50% 9,58% 23,75% 6,67% 3rd option 30,42% 20,83% 14,58% 15,00% 11,67% 3rd option 7,50% 31,25% 46,25% - Table 6. Disadvantages of road pricing scheme 1st option 2nd option User cost Loss of privacy Infrastructure cost No disadvantage 68,75% 15,00% 10,42% 5,83% 17,92% 41,67% 32,50% - ISBN: 978-960-6766-87-9 81 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 Table 7. Use of revenues 1st option Environment improvement 24,58% Public transport improvement 32,50% Social Causes 19,58% Road network improvement 14,58% Lower taxes 8,75% 2nd option 22,92% 25,42% 11,67% 32,92% 7,08% 3rd option 22,08% 16,25% 20,83% 24,17% 16,25% Coefficient 0,385 0,330 0,810 -0.679 1,936 -2,231 4th-5th option 28,33% 24,17% 42,50% 26,25% 63,33% t-stat 1,995 27,500 2,596 -1.656 1,825 -2,979 not selected 2,08% 1,67% 5,42% 2,08% 4,58% Significant at 95,4% 95% 99% 90.3% 93,2% 99,7% Table 8. Driver acceptability factors Trip frequency Trip duration Driver Perception: Revenue use on public transport Driver Perception: Revenue use for lower taxes Driver Perception: No disadvantage of congestion pricing Constant Table 9. Driver willingness to pay factors Coefficient Difference between travel time reduction of the chosen route and the non-chosen route Difference between travel time deviation of the chosen route and the non-chosen route Road pricing acceptability Driver Perception: Revenue use for lower taxes Driver Perception: Advantage of road pricing Reduction in congestion Profession Constant 14,424 -10,907 ,289 -,675 ,408 -,324 -1,818 t-stat 16,714 -16,121 1,622 -2,857 2,459 -1,465 -9,562 Significant at 100,0% 100,0% 89,5% 99,6% 98,6% 85,7% 100,0% ISBN: 978-960-6766-87-9 82 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 Analyzing the types of urban physical development deviation using the structure of PERT technique, a glance to biology Wahid Yeganegi Dastgerdi MA of Urban and Regional planning fromUniversity of Tehran, Iran wyeganegi@yahoo.com Abstract: There are three different stages in physical development of each wholeness including growth, morphogenesis and differentiation. Since cities are physical entities, this three leveled process can be evaluated within all of them. It is interesting but the important question is that how we can monitor this process so that none of the details are neglected. Here the PERT (program evaluation and review technique) can be very useful. It defines two factors: process and activity, and using these two factors, planners can monitor and evaluate all detailed incidents during the physical planning of the cities. Therefore the planner can analyze type and the time of beginning of deviations. Keywords: urban physical development evaluation, deviation, growth, morphogenesis, differentiation, PERT technique 1. Introduction Urban physical development is a vital subject for city managers, local authorities and even peoples. It is important for city managers and local authorities because they intend to know hoe much their city has capacity to grow and how they must allocate fiscal recourses to each part of the city. The importance of it to people is because they want to know the optimal place to invest. But it is hard to decide while we have not an obvious image from the physical development stage in each part of the city. Here using the structure of the PERT technique will help us to simplify the evaluation. famous persons who proceed the theory of main structure in his book " the new theory of urban design". Synchronous with him and after him many other people were interested to work on this field. Intention of this paper is to use this theory as a base such that can use biologists' theory of development and then with combining these two theories trying to analyze the quality of deviation through the life time of a city. 3. Concept of city's main structure Main structure theory says each city has two parts, a main structure that includes a hierarchy of centers, the routes between centers etc that forms the skeleton of the city and the other parts of it. 2. Background and objective: research about the physical development of the cities come back to the theories which are about the role of main structure .Christopher Alexander is one of the most ISBN: 978-960-6766-87-9 83 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 This theory also says we can effect on the whole of city by changing this structure. (Bahraini, 1999, p 15). This collection of series and parallel tasks can be modeled as a network. The Program Evaluation and Review Technique (PERT) is a network model that allows for randomness in activity completion times. PERT was developed in the late 1950's for the U.S. Navy's Polaris project having thousands of contractors. It has the potential to reduce both the time and cost required to complete a project. In a project, an activity is a task that must be performed and an event is a milestone marking the completion of one or more activities. Before an activity can begin, all of its predecessor activities must be completed. Project network models represent activities and milestones by arcs and nodes. PERT originally was an activity on arc network, in which the activities are represented on the lines and milestones on the nodes. Over time, some people began to use PERT as an activity on node network. For this discussion, we will use the original form of activity on arc. The PERT chart may have multiple pages with many sub-tasks.2 4. Physical developments in biology Biologists recognized three stages during the physical development process in living creatures, including growth morphogenesis and differentiation. As a simple definition growth means increase on quantity whereas development is related to qualitative dimensions. Morphogenesis also is an amazing phenomenon. During this stage, form of a developing system emerges.1 Frankly we can say that morphogenesis is the architecture of physical development. Morphogenesis itself has two sub stages. At first the volume of the structure is increased by increase in size, number or distance of components or a combination of these three methods. This sub stage not happening randomly in every direction. It emerges from directions that improve the physical and structural balance. 5. PERT method Complex projects require a series of activities, some of which must be performed sequentially and others that can be performed in parallel with other activities. - Britannica encyclopedia 6. Evaluation of deviation of physical development through the time life of a city Since the physical development itself is a common concept we can use this classification in urban physical development field. 1 2 -www.netmba.com ISBN: 978-960-6766-87-9 84 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 So growth is the first stage of any urban physical development. It emerges from directions that improve the physical and structural balance. If we present the process of physical development, will have a chart like below. Physical development according to biologist theories (Britannica encyclopedia) Now the structure of PERT (program evaluation and review technique) can help us to evaluate and monitor the physical development of different districts of each city. In PERT technique, physical development of a city is equal with a work , While the products and patterns of a stage transforms to products and patterns of consequence stage we encounter to a process and activity in PERT is equal to initiation or stopping point of each stage. So according to the chart, deviation can happen during each three activities or each two processes of 1 or 2. So in the chart above, growth, morphogenesis and differentiation are equal to activities and 1&2 are equal to processes that are between these activities. Now we can classify the type and quality of urban physical development deviations. The table below presents the type and quality of these deviations: Table no. 1 Deviation type A or deviation through growth: As we defined growth formerly, it means adding new components to the previous ones of wholeness. So here the quality is not so important. Deviation in growth includes shortage or overstock of similar components in a part of city. When in these districts we upload (population, land use etc.)more than infrastructure capacity will encounter to this type of deviation, Deviation type B or deviation in morphogenesis activity: main structure of a city has a great role in shaping the wholeness of a city. During the morphogenesis activity, there will be generate products and patterns which are city centers and patterns of their distribution and interrelationships at city. So here the agents of deviation are those who affect the location of centers, importance of each center among their hierarchy and quality of interrelationships of centers in a city. Deviation type C or deviation in differentiation activity: This type of deviation occurs when land uses are not applicable to their bed's situations or there is no coordination between the positions the land use and its location in relation with main structure. Deviation during "1" process : ISBN: 978-960-6766-87-9 85 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 This type of deviation occurs while we don’t allocate resources to different part of a city according to the time table of city physical development that result in macrosofaly. Deviation during "2" process: This type of deviation occurs while the main structure of a city is formed but the centers haven’t been specialized. Conclusion: Based on the article we can define indices to measure the quality and quantity of the physical development of a city. This method can help urban designers and planners and even local authorities to develop the city much better. Here we represent that there can be five different classes of deviation through the lifetime of city. Defining the indices also needs a different research but it can be a new horizon to the physical development of the cities. REFRENCES: 1- Bahraini, Seyed Hosein, Urban design process, 1999, University of Tehran press, Iran. 2- Britannica encyclopedia, Britannica encyclopedia Ltd, London, 1997. 3- www.netmba.com ISBN: 978-960-6766-87-9 86 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 CUSTOMER SATISFACTION OF LIGHT RAIL TRANSIT (LRT) Ng Ying Peng, Wan Rosmanira Ismail, Nur Riza Suradi, Zainol Mustafa, Zalina Mohd Ali, Rofizah Mohammad & Faridatulazna Ahmad School of Mathematical Sciences, Faculty of Science & Technology, National University of Malaysia. Abstract: The concept of customer satisfaction is important in any business. Striving for customer satisfaction means understanding and anticipating what the customers want of the products or services given. Customer satisfaction can also be taken as an indicator either a customer will buy or use the product continuously. The aim of introducing the Light Rail Transit (LRT) system in Malaysia is to ease the traffic congestion in Malaysian roads by providing alternative mode of transport. If the service is not satisfactory and customers are not keen of using the service then the purpose of introducing the service in the first place is defeated. Due to that reason we wish to study the current level of customer satisfaction towards LRT service. From this study, the service attributes that need to be improved in order to increase the level of customer satisfaction will be identified. The service atributes are classified into two categories attributes at the train stations and attributes inside the train. A customer satisfaction survey were carried out and 300 questionnaires were successfully completed by the respondents. The data is analyze using the Penalty-Reward Contrast Analysis (PRCA) and Adequacy Importance Model (AIM). PRCA is used to classify the service attributes and AIM is used to measure the overall satisfaction of each service attributes by taking both the importance and satisfaction of each attributes into account. The resulting classifications of service attributes will be illustrated using Kano Model. The results show almost all service attributes are at a satisfactory level except four. These attributes were identified as cleanliness, efficiency, service of staff and safety. Further improvement need to be taken to improve the service and to increase the level of customer satisfaction. Keywords: customer satisfaction; service, service attributes, PRCA , AIM, Kano model 1. Introduction The concept of customer satisfaction is important in business field. Striving for customer satisfaction means understanding and anticipating what customers want of the products in the future but do not expect of them (Matzler et al. 1996). Customer satisfaction can be taken as an indicator that either customer will buy or use this product continuously. The product is indicated as quality if it success to achieve the customer expectation and needs. On the other hand, the failure to achieve customer satisfaction on this product will lose their ability in market competitor. For service providers, customer satisfaction is crucial to know which service attributes add value and increase satisfaction, which of them merely fulfil minimum requirements and minimum dissatisfactions and which do both. Only then can they make better decisions about the research should be allocated to different service attributes in order to improve quality and satisfaction. Thus, the identification of customer satisfaction factors is crucial (Matzler & Sauerwein 2002). ISBN: 978-960-6766-87-9 87 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 In the transportation of Light Rail Transit (LRT), customer satisfaction concept is also crucial which service quality is indicated by customer evaluations (Sek 2002). In fact, rail system is a high technology transportation which is providing alternatives yet comfortable for customer. In Kuala Lumpur (KL), there are two LRT system implemented which are namely as Kelana Jaya Line (before that is PUTRA-LRT) and Ampang line (before that is STAR-LRT). The distance connection of LRT system between city central and other districts is around 20 km radius and it was a crossing with total 49 stations. The stations are styled in several types of architectural designs. Elevated stations, in most parts, were constructed in four major styles with distinctive roof designs for specific portions of the line. Underground stations, however, tend to feature unique concourse layout and vestibules, and feature floor-to-ceiling platform screen doors to prevent platform-to-track intrusions. Stations with island platforms allow easy interchange between north-bound and south-bound trains without requiring one to walk down/up to the concourse level. General speaking, LRT system play a role in reducing the transport crowdedness on the road. Nevertheless, issues and customer complaints of LRT system problem are increasingly in several years. According to the The Sun (2001) report, one Argentina passenger fallen suddenly on the track at Station Bangsar. If not have the Platform Intrusion Emergency Stop (PIES) system to stop the PUTRA-LRT train, the passenger may threaten by this an accident. Besides that, a STAR-LRT train overshot at the ends of its track and daggling around 30m heights at Sentul Timur Station, Kuala Lumpur. Pondering whether did the LRT Company really service or maintain their train regularly to ensure public safety as this is not the first time accident (Ong 2006). Moreover, people crowdedness at station LRT, especially at Kuala Lumpur City Center (KLCC) station was caused of the lack of LRT train frequency during peak hour (Shek 2007). According to Berita Harian (2007), increasing of the Monthly Travel Card (MTC) for the LRT fare also was complained by some people, which the LRT Company just to emphasize on increasing the LRT fare without give a notice early to passenger. Base on those problems above, LRT Company should to take seriously on improving of LRT service in order to increase customer satisfaction and service quality. The objective of this study is to investigate the level of customer satisfaction on the LRT service which is providing by LRT Company. Furtheremore, this study will also identify service dimension which is needed to improve in order to increase the level of customer satisfaction. The service dimensions are classified into at station and inside the train. Data analysis has been done through Penalty-Reward Contrast Analysis, PRCA and Adequacy Importance Model, IAM and following with illustrated all service dimensions in Kano Model. 2. Methodologi The survey questionnaires in this study were developed based on service dimensions such as cleanliness, efficiency, facility, comfort, safety, information delivery, price and staff service. All passengers who always to use service LRT are become respondents for this study. Around 300 respondents was conducted and successfully to answer the whole questionnaires. 3. Kano Model Kano analysis is indicated as a quality tool used to identify which customer needs is important. The requirement of all customers are may not same as important. Therefore, Kano analysis has a vital role to play in arranging the different customers requirements and identify which requirement is most important for all customers. Kano Model (Figure 1) distinguishes between three types of product requirement which influence customer satisfaction in different ways when met (Matzler et al. 1996). Three types of product requirement can be classified as basic factor, performance factor and excitement factor. ISBN: 978-960-6766-87-9 88 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 Figure 1: Illustration of Kano Model Excitement factor can be described as surprise and delight attributes, and provide satisfaction when achieved fully but do not cause dissatisfaction when not fulfilled (Kano et al. 1984). Normally, those requirements are neither explicitly expressed nor expected by the customer (Matzler et al. 1996). Since this type of requirements often unexpectedly delights customers, they are often unspoken (Witell & Lofgren 2007). Performance factor result in satisfaction when fulfilled and result in dissatisfaction when not fulfilled (Kano et al. 1984). With regard to these requirements, customer satisfaction is proportional to the level of fulfilment which the higher the level of fulfilment, the higher the customer’s satisfaction and vice versa. Performance factor is usually explicitly demanded by the customer (Matzler et al. 1996). Basic factor is taken for granted when fulfilled but result in dissatisfaction when not fulfilled (Kano et al. 1984). Explicitly, customer regards those requirements as prerequisites. If these requirements are not fulfilled, the customer will be extremely dissatisfied. On the other hand, as the customer takes these requirements for granted, their fulfilment will not increase his satisfaction. Fulfilling the basic requirements will only lead to a state of “not dissatisfied” (Matzler et al. 1996). However, some requirements of products cannot be classified according to the Kano Model. These requirements are neither good nor bad, and consequently they do not result in either customer satisfaction or customer dissatisfaction. As conclusion, a competitive product meets the requirement of basic factor, maximises performances factor, and includes as many “excitement” factor as possible at a cost the market can bear. 4. Penalty-Reward Contract Analysis (PRCA) In this study, Penalty-Reward Contrast Analysis (PRCA) was applied which suggested by Brandt (1987). This method has two advantages. First, the attribute importance given the level of attribute satisfaction can be measured. Second, ISBN: 978-960-6766-87-9 89 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 the results allow classifying service attributes into basic, performance and excitement factors. Hence, the convergent validity can be assessed (Matzler & Sauerwein 2002). Brandt (1987) proposes a type of regression analysis that uses dummy-variables to identify minimum requirements (basic factors) and value-enhancing requirements (excitement factors). In essence, one set of dummy variables is created and used to quantify excitement factors and another set is created to quantify basic factors. Basic factors and excitement factors are expressed as contributor to overall satisfaction (dependent variable) (Matzler & Sauerwein 2002). Generally, the code of dummy-variable can be explicated in Table 1. Table 1: Attribute satisfaction rating change to dummy-variable Customer satisfaction Very satisfied Moderate Very unsatisfied D1 D2 and dummy-variable for rewards and penalty as independent variables. If the reward outweighs the penalty, the attribute is considered as an excitement factor. If the penalty exceeds the reward, the attribute in question is a basic factor. If reward and penalty are equal, the attribute lend to satisfaction when performance is high as well as to dissatisfaction when performance is low. Hence, it is a performance factor “hybrid” (Matzler & Sauerwein 2002). Table 2 as below indicated the condition of categorization attribute into basic factor, performance factor and excitement factor base on PRCA method. Table 2: Condition of categorization attribute Reward Penalty Factor Significant Insignificant Excitement Significant Significant Performance Insignificant Significant Basic Insignificant Insignificant Indifferent The value of model significant is indicated by comparison of model value-p with fixed value-α. If value-p is lower than value-α, this attribute is indicated as significant. On the contrary, the attribute is indicated as insignificant if value-p is higher than value-α. Next, the attribute can be classified whether as basic factor, performance factor, excitement factor or indifferent by comparison the significant of both reward and penalty coefficient. 1 0 0 0 0 1 D1 = “Reward” are expressed as incremental increase associated with high satisfaction D2 = “Penalty” are expressed as incremental decrease associated with low satisfaction In order to conduct the analysis in this study, attribute satisfaction ratings were recorded. “Very satisfied” ratings were used to form the dummy variables to quantify excitement factors (value of “0”), while “very unsatisfied” ratings were used to form the dummy variables to quantify basic factors (value of “1”). “Neither satisfied nor unsatisfied” were defined as expressing indifferent (expectations were met). Indifferent customers comprise a reference group (Matzler & Sauerwein 2002). Base on this recording, a regression analysis is conducted by using seven-point overall satisfaction ratings as dependent variable 5. Adequacy Importance Model (AIM) Adequacy Importance Model is suggested by Dobbelstein (2003) which is used to measures the satisfaction in a compensatory way. As an example, a low product value for one requirement (Customer A, aspect of frequency of LRT train) can be compensated by a high product value for another requirement (Customer A, aspect of accuracy time arrival of LRT train). This indicates that customer satisfaction on LRT service in efficiency dimension will not be totally lost if it performs poorly in one requirement as ISBN: 978-960-6766-87-9 90 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 long as there are other good ones. The overallsatisfaction scores for efficiency dimension can be obtained by the average of all customers’ satisfaction on this dimension. Below equation has shown the calculation on the overallsatisfaction for each aspect of product Pijk = n for customer i Perceived quality of product attribute k of dimension j by customer i = Number of relevent product attributes. n Sij = k=1 ∑ Iijk * Pijk Sij Iijk = Overall saisfaction of customer i with dimension j = Importance of product attribute k The use of the Adequacy Importance Model is explained with the help of an example in Figure 2. It gives an example of the scales that may be used and of a concrete measurement and calculation of the overall satisfaction of three customers LRT. In the example customer A is most satisfied with the performance of efficiency dimension. Example of Using the Adequacy Importance Model to Measure a Customer Satisfaction on LRT Service Extremely Unsatisfied How satisfied are you with accuracy time arrival of LRT train as display on digital board? Extremely Satisfied 1 2 3 4 5 6 7 Extremely Unsatisfied How important are you with accuracy time arrival of LRT train as display on digital board? Extremely Satisfied 1 2 3 4 5 6 7 Customer A B C Accuracy time arrival of LRT train as display on digital board Pijk Iijk 5 7 4 5 3 7 Frequency of LRT train Pijk 2 4 3 Iijk 6 6 5 Overall satisfaction in efficiency dimension Sij (5*7) + (2*6) = 47 (4*5) + (4*6) = 44 (3*7) + (3*5) = 36 Figure 2: Example of Adequacy Importance Model ISBN: 978-960-6766-87-9 91 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 6. Disscussion of Result and Finding The overall satisfaction of each service dimensions from the result of IAM are scheduled in Table 3 and 4, which the service dimensions are classified into two sections. First section (Table 3) is about satisfaction mean and standard deviation of service dimensions at station while second section (Table 4) is about satisfaction mean and standard deviation of service dimensions inside the train. Table 3: Satisfaction mean and standard deviation of service dimensions at station Service Dimension Cleanliness Efficiency Facility Comfort Information delivery Price Staff service Mean 22.89 24.75 26.01 26.72 28.68 28.27 25.18 Standard Deviation 11.556 10.834 8.703 11.794 8.327 12.505 9.353 the overall satisfaction which are lower than value of 25, it means those service dimensions are classified into dissatisfied group. While if the overall satisfaction of service dimensions are higher than value of 25, it means the service dimensions are classified into satisfied group. Dissatisfied groups are needed to take an action to improve their service while satifed groups are indicated as maintain or to upgrade their service which can to attain a higher level of satisfation. Based on the Table 3, service dimensions at station are classified as dissatisfied group including cleanliness and efficiency. Staff service is indicated as medium satisfied bacause it’s satisfaction mean is nearly to neutral value and it should to get an improvement as possible. On the other hand, only safety is needed to get an improvement among others service dimensions inside the train (Table 4). 6.1. The combination result of PRCA and IAM The result of combination between PRCA and IAM has shown in Table 5. As a result, the majoriti of service dimensions are classified into satified group excluded cleanliness, efficiency and staff service dimension at station as well as safety dimension inside the train. Safety and clenliness are needed to take a seriously for improving the service quality because both service dimensions are classified as basic factor with a lower satisfied. Both service dimensions are regarded as prerequisites by LRT customers and customers will be extremely disappointed and dissatisfied if this factor cannot be fulfilled. Besides cleanliness and safety dimension, efficiency and staff service are classified as performance factor with lower satisfied also need to be focused. Both service dimensions will lend to dissatisfaction when the service quality cannot be fulfilled as well as satisfaction when a higher fulfilment of both service quality dimensions. Table 4: Satisfaction mean and standard deviation of service dimensions inside the tren Service Dimension Cleanliness Efficiency Facility Safety Comfort Information delivery Mean 33.12 29.42 27.28 21.18 28.86 29.46 Standard Deviation 12.040 13.210 11.255 6.645 10.778 9.999 The value of 25 is indicated as neutral (The highest value is 49 while the value of 1 as the lowest). Thus, if the service dimensions have ISBN: 978-960-6766-87-9 92 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 Table 5: The combination result of PRCA and IAM Classification by PRCA Overall satisfaction of IAM Dimension At Station Cleanliness Efficiency Facility Comfort Information delivery Price Staff service Decision Basic Performance Performance Excitement Performance Excitement Performance Low Low High High High High Medium Improvement Improvement Persisting Persisting Persisting Persisting Improvement if possible Persisting Persisting Persisting Improvement Persisting Persisting Inside the Train Cleanliness Efficiency Facility Safety Comfort Information delivery Performance Basic Performance Basic Performance Performance High High High Low High High 7. Conclusion As a result of the present analysis on service dimensions LRT, it can be concluded by saying that customer satisfaction on LRT service just remain to medium satisfied level. The service dimensions are needed to focus included cleanliness, efficiency, staff service at station as well as safety inside the train. Figure 3 is the illustration of service dimensions at LRT station while Figure 4 is the illustration of service dimensions inside the LRT train. Based on the result of decision, LRT Company should be taken actions for improving the service quality in those dimensions which are classified in dissatisfied group in order to increase the service quality and customer satified. Total Quality Service (TQS) is an important concept which is used to improve the service quality. According to Stamatis 1996, TQS is indicated as a sincerely commitment to concept operation which focus on customer, increasing service performance, measuring performance company by using reference measurement, respect and provide a reward to staff, and to attain customer demand on anytime. TQS was suggested to implement on LRT Company because it is a good strategy system which can to integrate management system to involve all leader and company staff by using combination method of qualitative and quantitative for continually improvement process in order to meet customer needs. ISBN: 978-960-6766-87-9 93 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 Customer Satisfaction Delight Excitement Factor • Comfort • Price Performance Factor • Facility • Information Delivery Immediate Happiness Indifferent Poor Good Degree of Implementation Not Unhappy Performance Factor • Efficiency • Staff Service Basic Factor • Cleanliness Disappointed Table 3: The illustration of service dimensions at LRT station Customer Satisfaction Performance Factor Delight • Cleanliness • Facility • Comfort • Information Delivery Excitement Factor Immediate Happiness Poor Indifferent Good Degree of Implementation Not Unhappy Basic Factor • Efficiency Disappointed Basic Factor • Safety Table 4: The illustration of service dimensions inside the LRT train ISBN: 978-960-6766-87-9 94 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 Refrences Brandt, R. D. 1987. A procedure for identifying valueenhancing service components using customer satisfaction survey data, hlm. 35-41. Chicago: American Marketing Association. Berita Harian. 2007. 2 November. Dobbelstein, T. & Taylor, S. 2003. Analysing the world of work’s requirements with the aim of enthusing companies about cooperative education, Asia-Pacific. Journal of Cooperative Education. 5(1): 1-6. Kano, N., Seraku, N., Takahashi, F. & Tsjui, S. 1984. Attractive quality and must-be quality. Hinshitsu 14(2): 147-156. Matzler, K., Hinterhuber, H. H., Bailom, F. & Sauerwein, E. 1996. How to delight your customers. Journal of Product & Brand Management 5(2): 6-18. Matzler, K. & Sauerwein, E. 2002. The factor structure of customer satisfaction: An empirical test of the importance grid and the penalty-reward-contrast analysis. International Journal of Service Industry Management 13(4): 314-332. Ong, J. 2006. Malaysia LRT train overshoot. (http://www.malaysia_lrt_train_overshoot_files.) [1 April 2008]. Sek, S. K. 2002. Kajian kepuasan pelanggan dalam sektor service: satu kajian kes terhadap syarikat LRT Malaysia. Tesis Sarjana, Pusat Sains Matematik, Universiti Kebangsaan Malaysia. Shek, V. 2007. Light Rail transit (LRT) just too packed at peak hours. (http://blog.thestar.com.my/permalink.asp?id=10781) [12 Mac 2008]. Stamatis D. H. 1996. Total quality service. Principles, practice & implementation. USA: St. Lucie Press. The Sun. 2001. 16 Ogos Witell, L & Lofgren, M. 2007. Classification of quality attributes. Managing Service Quality 17(1): 54-73. ISBN: 978-960-6766-87-9 95 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 Complex Network Applications to the Infrastructure Systems: the Italian Airport Network case J. Quartieri*, M. Guida*, C. Guarnaccia*, S. D’Ambrosio**, D. Guadagnuolo* *Department of Physics “E. R. Caianiello” and Faculty of Engineering, University of Salerno, quartieri@unisa.it , miguida@unisa.it **Department of Mechanical Engineering and Faculty of Engineering , University of Salerno Abstract: - In this paper the structure of the Italian Airport Network (IAN) has been analysed, describing it as a mathematical graph and investigating its topological properties by means of a non standard point of view based on the approach of the multiple addendials. It turns out that this network has remarkable and interesting features, such as like the one to be a scale-free network, since both the degree and the betweenness centrality distributions follow a typical power-law behaviour, well known in literature like a Double Pareto Law. Moreover, these distributions exhibit a number of interesting peculiarities, like the existence of some “hubs” in the network, i.e., in the graph theory’s jargon, nodes with a very large number of links. According to a preliminary investigation the IAN can be also considered a candidate to represent a small-world network as the mean distance L between reachable pairs of airports grows at most as the logarithm of the number of airports. Key-Words: - Airport Network, Graph, Multiple Addendials. 1 Some mathematical background With the aim to ease reading of this paper also to general people we report some useful considerations on some aspects of the mathematical approach to complex networks. This goal will be accomplished utilizing a non standard point of view, whose basic formulation has been already introduced in past papers by one of us [1]. ∑ n |k = n |k +1 n (2b) we immediately have: ∑ n |k = ( n + 1 ) |k k (4) ∆ n |k = n |k −1 n (5) 1.1 Multiple addendials Referring to [1] for more details, here we address just the main features of an “addendial”. An addendial of first degree is defined as the sum of the first integer numbers: n | 1 ≡ n + ( n − 1 ) + ... + 2 + 1 = n ( n + 1 )  n + 1 =  2   2   ∆ n |k = ( n − 1 ) |k k (6) (1) Being ∆ f j ≡ f j − f j −1 , Σ turns out to be just its inverse j j An addendial of “k”-th degree is, therefore, the sum of the first addendials of “k-1”-st degree: operator. In fact ∆ Σ = Σ ∆ = I j j j j (7) n + k  n |k = n |k −1 +( n −1 ) |k −1 +... + 2 |k −1 +1|k −1 =   k +1    In a short notation: (2) 1.2 Maximum number of links Let us, now, consider a network consisting of n distinct points (n-vertices or n-nodes) between which we can establish connections (edges or links) in such a way that any two points are connected by links (if each pair of nodes is connected by one and only one link, the network is named “simple”). In this work we deal only with simple networks. The points connected each other in this way will be called “neighbours”. Such networks can be mathematically represented by graphs. Let us calculate how many links can be drawn among n assigned nodes: M(n). n | k = ∑ n |k −1 n (2a) This relation together with the definition (1) suggests to put: n ≡ n |0 . (3) Multiple addendials are integer functions of two integer variables and have very remarkable properties [1]. Besides (2a), which we can rewrite as: ISBN: 978-960-6766-87-9 96 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 Fig. 1a and Fig. 1b Then, suppose we already have (n-1) connected nodes (Fig. 1a). Among them we have M(n-1) links. If we add one more node we have to insert (n-1) new links (Fig. 1b). In formula: M ( n ) = M ( n −1) + ( n −1 ) ; M ( n ) − M ( n −1 ) = ( n −1 ) ; Applying the operator Σ to both sides: Before, we saw that: M ( n ) ≡ M ( n ,1 ) = ( n − 1 ) | 1 ≡ X Appling the “zero rule” for independent procedures, we can say: in a set of n-nodes, with no link, we can insert the first link in X different ways. Having placed the first link, we can put the second one in (X-1) ways. So, if the two links were distinguishable we would have M ( n ,2 )dist = X ( X − 1 ) , but, as we are interested in studying indistinguishable links, we have, theferore, to divide by 2! (permutations): M ( n ,2 ) = ∆ M ( n ) = ( n −1 ) X ( X −1) . 2! Iterating, we have: Σ∆ M ( n ) = Σ ( n − 1 ) M ( n,s ) = Simplifying the two inverse operators in the l.h.s., and recalling (3) and (2b), we finally have: M ( n ) = ( n − 1 ) | 1 + const In order to determine the value of the constant we impose the initial value condition: f ( 2 ) = 1 , which means that between only two nodes we can draw just one link, and we have: X ( X − 1 )...( X − s + 1 )  X  =   and, finally: s s!    (n − 1) |1   M ( n, s ) =   s    Let us now introduce few characteristic features useful in the study of complex networks. Adjacency matrix A square matrix A(n × n ) , whose elements aij take value 1 if there is a link from node i to node j and take value 0 otherwise, is termed the adjacency matrix. For the example below it is: 1 = M ( 2 ) = 1 | 1 + const = 1 + const → const = 0 So M ( n ) = ( n − 1 ) | 1 which also means:  n − 1 ( n − 1 ) n M( n ) =  = ( n − 1 ) + ... + 2 + 1  2 =  2   For future reference we put ( n − 1 ) | 1 ≡ X 1.3 Maximum number of s-links Now we want to extend this analysis to a wider class of mathematical objects. Let us begin calculating the maximum number of 2-links M ( n ,2 ) . We call “2-link” a sub-graph with only two links (in sequence or not) in an n-nodes graph: In sequence: Not in sequence: As each link is not directed (undirected network) the matrix A turns out to be naturally symmetric. Connection degree The connection degree ki of a node i is the number of other nodes to which it is connected, or, shortly, the number of its neighbours: k i = ∑ aij j =1 n ISBN: 978-960-6766-87-9 97 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 Distance matrix The distance matrix D(n × n ) has as elements {dij }, which are the number of steps, along the shortest path, linking every pair of connected nodes. The mean distance of a connected network is defined as: Lconn = 1 n ∑ d ij 2 i , j =1 # pairs = 1 n ∑ d ij 2 i , j =1 X conn . Clustering coefficient the clustering coefficient of node i, Ci , Ci ∈ [0,1] , can be defined as the following ratio, where : the numerator is equal to the number of pairs made of neighbours of i which are themselves neighbours, i.e., neighbours of each other or equivalenty themselves connected by a link. the denominator is equal to the number of all the possible pairs of neighbours that could be in principle constructed from the ki neighbours of node i, i.e. (ki − 1) |1 : The last equality holds because # pairs equals the maximum number of links, X conn . When X conn is written in terms of the number of connected nodes, nconn , we have: Lconn = n 1 ∑ d ij nconn ( nconn − 1 ) i , j =1 and Lconn is also named the characteristic distance. For the previous example we obtain: k 2 = 3 k3 = 3 k 4 = 3 L conn = 35 20 k5 = 3 k6 = 2 C 2 = 2 / 3 C3 = 1 / 3 C4 = 2 / 3 C5 = 1 / 3 C 6 = 0 2 The Italian Airport Network We investigated the topological properties of the Italian Airport Network (IAN), representing it like a mathematical graph where to each airport is associated a node and pairs of nodes that are just connected by non-stop passenger flights are linked together. In order to accomplish that, we have studied the data derived from the OAG Max database [2], compiled by OAG Worldwide (Downers Grove, IL) including all the scheduled flights and scheduled charter flights of the world’s airlines both for big aircrafts (air carriers) and small aircrafts (air taxis) for the period June 1, 2005, to May 31, 2006 (period (o)). At the end of this construction, whose full details are contained in a previous paper by one of us [3], the data corresponding to IAN have been extracted and the resulting directed graph has turned out to be made of 42 nodes and a total number of links (non-stop flights) equal to 310 (Fig. 2). Following the definitions given in sect. 1, let us consider the adjacency matrix, A(n × n ) , whose elements aij , in the case of the IAN, take value 1 if on any day of the week there is a flight from node i to node j by any service provider and 0 otherwise. In order to simplify the following analysis we will restrict ourselves just to data pertaining to flights available in the period (o). Similar investigations related to other periods of time are described in details in [3]. The diameter D of the network can be defined as the length of the maximum “shortest path”. D = max( d ij ) . In the previous example D=3. Betweennes or centrality The betweenness or centrality Bi of node i is defined as the number of all “shortest paths” linking any two different nodes and passing through i. . e.g. B2 = 5 , B6 = 0 , B3 = 3 We define the normalized betweenness bi = Bi / B , where network. B represents the average betweeness of the ISBN: 978-960-6766-87-9 98 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 Fig.2: The Italian Airport Network (IAN) Then, we calculate for each node i the connection degree ki , that is the number of other airports to which airport i is connected by a non-stop flight, and the corresponding cumulative distribution cumulative distribution P (> k ) , which gives the probability that a node has k or more connections to other nodes and it is defined as: P(> k ) = ∑ p (k ′) , where p(k ) is the probability density, k ′= k ∞ where kc = 9 and (α1 , α 2 ) = (0.2, 1.7) . This behaviour is typical of many complex networks, termed “scale-free” [5]. The fact that the exponent α 2 of the degree distribution is greater than one suggest the possibility that the structure of our network is fractal as proposed in [6]. The fact that networks are scale-free was shown to have important implications on network robustness to random failures and vulnerability to attack [7]. The distribution reportedin Fig. 3 clearly shows the existence in the Italian Airport Network of so-called “hubs”, i.e. nodes with a large number of connections: in our case Fiumicino (FCO), Olbia (OLB), Milano-Malpensa (MXP) and Catania (CTA). The degree of a node is a importance source on its own but, however, does not provide complete information on the role played by the single nodes in the network. To detail a bit more the role of each node inside the network let us adapt, according to the definitions already given in sect. 1, the so-called “betweenness centrality” of cities, introduced in [8-9]. Besides the betweenness Bi of each airport i it is convenient, as done in [3, 10], to introduce the corresponding normalized betweenness bi = Bi / B , where p(k ') = k ' / ∑ ki 1 n B represents the average betweeness of the network and compute the cumulative distribution P (> b ) as On a log-log scale the normalized cumulative distribution P(> k ) versus the degree k looks like (Fig. 3). P(> b ) = ∑ p (b′) , where p(b ) is the probability density. For the period considered, we plot, on a log-log scale, the normalized cumulative distribution P (> b ) versus the normalized betweenness b (Fig. 4). As for the degree distribution (Fig. 3) the normalized cumulative distribution P (> b ) follows a Double Pareto Law. This behaviour turns out to be again typical of a scale-free network [4]. b′ = b ∞ The data distribution suggest that P (> k ) follows a Double Pareto Law [4] Fig.3: The normalized cumulative distribution P(> k ) versus the degree k for period (o). k −α1 for k ≤ kc  , P(> k ) ∝  −α k 2 for k > kc  Fig.4: The normalized cumulative distribution P(> b ) versus the normalized betweenness b for period (o). ISBN: 978-960-6766-87-9 99 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 Another property of the IAN network is that it is a smallworld network. According to Watts and Strogatz’ [11-12] a network is a small-world network when the mean distance among reachable pairs of nodes L grows at most as log e n , where n is the number of nodes. In order to check that, we calculate L , as it has been defined previously. For the IAN, it turns out to be that L = 1.97 and log e n = 3.74 for period (o). These figures indicate that L is closer to 3.74 rather than to the number of nodes, n = 42 and show that according to the definition [11-12], the Italian Airport Network is a “small-world” network. Another quantity that is worth to check in order to characterize the topological properties of a network, is the so-called network’s clustering coefficient C, defined previously. If we calculate C, according to the definition, for the IAN we find that, for the period (o) it turns out to be: C = 0.10 . References: [1] Quartieri J., “Some mathematical considerations arising in the study of a Degenerate Harmonic Oscillator”, Lett. Nuovo Cimento, Vol. 35, N. 14, 1982, p.p. 433-436. [2] http://oagdata.com/solutions/max.aspx. [3] Guida M. and Funaro M., “Topology of the Italian Airport Network”, Chaos, Solitons & Fractals, Vol. 31, 2007, p.p. 527-536. [4] Li W., Cai X., Phys. Rev. E , Vol. 69, 2004, 046106. [5] Bàrabasi A. L. and Albert R., Science, Vol. 286, 1999, p.p. 509-512. [6] Song C., Havlin S. and Makse H.A. (2005), “Fractal growth of complex networks : repulsion between hubs”. ArXiv: cond-mat/0507216 v1. [7] Crucitti P., Latora V. and Marchiori M., “Model for cascading failures in complex networks”, Phys. Rev. E , Vol. 69, 2004, 045104(R). [8] Freeman, L. C. (1977), Sociometry, Vol. 40, 35–41. [9] Newman, M. E. J. (2001), Phys. Rev. E Stat. Nonlin. Soft Matter Phys., Vol. 64, 016131, 016132. [10] Guimerà R., et al. (2005), “The worldwide air transportation network: Anomalous centrality, community structure, and cities’ global roles”, Proc. of the Natl. Acad. of Sci. USA, Vol. 102, No. 22, 77947799. [11] Watts D.J., Strogatz S.H. (1998) ”Collective dynamics of 'small-world' networks”, Nature, Vol. 393, 440-442. [12] Watts D.J. (1999), “Small Worlds - The Dynamics of Networks Between Order and Randomness”. Princeton University Press. 4 Conclusion In this paper the topological properties of the system of the Italian airports have been investigated from a non standard approach to the theory of complex networks. The associated network, the so-called Italian Airport Network (IAN) have been constructed associating a node to each airport and an edge to each non-stop connecting passenger flight operating between two different airports in any of the seven days of the week. The topological properties of the resulting network have been carefully examined leading to the evidence of a scalefree behaviour in the degree of connection distribution, in which the existence of “hubs”, i.e. nodes with a high value of the degree of connection, clearly manifests. However, the scale-free behaviour turned out to be a little bit different from the ones already reported in the literature for other air transportation networks, suggesting that the formation mechanism model underlying the IAN could be pretty different from the ones proposed so far. The topological properties of IAN have been investigated and confirmed (e.g., the scale-free characteristics) considering the data available in different period of time related to different seasons of the year. In these cases the well-known tourist vocation of some Italian locations really makes the difference. From such a kind of analysis some interesting and precious results for professional operators of the National Airport Infrastructure can be easily extracted. Furthermore, IAN satisfies all the requirements prescribed by the definition of small-world networks. ISBN: 978-960-6766-87-9 100 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 Importance of Phase Changes in Balanced Growth of a City Wahid Yeganegi Dastgerdi wyeganegi@yahoo.com Abstract: Form of a city has an important role in how much it can be large and how much will be the expenses of it. So selection of the form can help any city be more successful. Many famous people have theories to explain which form is better but all of these theories are abstract. Christopher Alexander in his famous book, the new theory of urban design [1] referenced to this kind of growth but he didn’t explain it obviously. In this paper, this kind of growth and its benefits are explained. One of the most important thing in this kind of growth is phase changes (a point at which system properties change suddenly)[6]. Here using phase changes means how can a city growth without encountering with problems caused by growing up around the initial centre. Keywords: City Form, Phase Changes, Balanced Growth 1 INTRODUCTION 2 BACKGROUND AND OBJECTIVE Designing the cities is a very importance matter for any government because the cities of any country can show the condition and ideology of that government. From the first civilizations in Egypt and Iraq in east to the last civilization in west all of them like to show their power and importance through their cities [4]. Although modern life in the contemporary cities is an additional factor which forces the governments design the cities. On the other hand we can find roles which exist in the nature as the best solution for growth. In continue it will be shown how this regulation can help urban designers make cities more efficient. Urban design became more important when the cars dominated on the cities. They need long straight streets and force cities grow up more and more and so urban design field was defined to solve the problem. From the earliest years of twenty century many different methods were presented by urban designer which try to find a good solution for problem of city growth. Taking a look to the Lynch's book, "the good city form theory" we can see the most popular theories presented up to now [3]. Any contemporary city has grown due to the one or mixture of some of these methods. Some of the must famous methods are presented here. Figure 1 : left, linear city presented by Imata. Right, radial-ring shaped city presented by Howard [5] ISBN: 978-960-6766-87-9 101 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 Figure 2: left, city in height presented by Le Corbusier, right, neighborhood unit presented by Stein[5] By passing time problems goes back to cities. Linear cities couldn't be proper for any situation because by growing the city, its parts became so far from each other [5]. Radial-ring shaped cities couldn't be the fine solution because when city growths a large volume of traffic goes round the city centre. City in height theory which Le Corbusier was one of the most famous fans became a very popular solution for many cities like NY and Tokyo although French didn't go along with it [5]. It could not be the proper solution for any city. Only cities in a developed country can growth due to this method because it needs a very developed transportation infrastructure. Stein theory is an amazing one which didn’t perceive any time. It was used to develop a small sub urban area around New York only [5]. It will be perceived the goodness of this theory in the last part of this paper. He defined some areas as neighbourhood areas based on centrality of initial services and primary schools. The first area had 0.5 mile radius and the second area had 1 mile radius. Centre of the second area is exactly settled on the edge of first area. Scrutinizing the above examples every one can perceive that by growing up the city during the time pressure on the city centre become more. 3 GROWTH IN THE NATURE On the other hand in nature we encounter with a different type of growth method. In this method at first tissue growing up around the initial centre after a while when the growth around the first centre cause many problems for the tissue a new bigger centre emerges on the edge of initial growth area(phase changes)[6]. The second centre becomes the centre of wholeness and this process can repeat some time. ISBN: 978-960-6766-87-9 102 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 Figure 3- in nature a wholeness don’t grow up around a fix centre but there is a point at which system properties change suddenly. Here it is presented how clay becomes dry and how tree bark forms. The picture below represents how this regulation can form wholeness. Next pictures show this method of growth. Figure 4- how a wholeness grow up in nature due to the self organization laws[5]. At first the wholeness grow up around the initial centre(steps 1 and 2) and then a new centre emerges at the edge of initial area of development then after a while a new and bigger centre emerges(step 3) and after that the wholeness will grow up around the new centre. Next picture has more explanation. Figure 5- how the nature model of growth cause balance between the initial development area and following expansion area. ISBN: 978-960-6766-87-9 103 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 4 IMPORTANCE OF PHASE CHANGES IN BALANCED GROWTH OF A CITY This kind of growth method has many benefits: 1. at first the pressure on the initial centre stops by emerging a new centre for the wholeness 2. traffic flow distributes symmetrically between the old area and new developed areas around the new centre 3. since the new centre has many choice to be settled on the edge of old developed area, city can grow up freely in every direction in concern with its conditions. 4. the city can have more than one centre at a time 5. in this method a hierarchy of centers exist which help the city to be more coherent. 6. if the radius of initial development area planned due to the distance which a pedestrian can travel easily(about 400 meter) we will encounter with a city which is modular in regard to pedestrian rather than cars. So here we can observe how close are Stein theory with the nature growth model. Although Stein theory is relatively week in explanation of how a city must grow up because he stopped at designing only a rather small district not planning a model to growth. This kind of growth in the nature is based on the Self Organization principles [6]. Unfortunately none the popular growth model in the urban design filed are based on what happens in the nature and all of them are abstract ideas. 5 CONCLUSION All of the theories in city growth field are only abstract ideas. It causes many problems for contemporary cities. Most popular problem are increasing compression in the city centers' area, need to a high expense to develop transportation infrastructures, unbalanced development of cities, traffic jam in the central part of cities and in the long run a highly polar city which absorb every resource to the centre. In the nature we encounter to a different method of growth which is close to the Stein theory of neighborhood but has many differences to it. The most important thing in this kind of growth is called phase changes which means a point at which system properties change suddenly, e.g. where a matrix goes from non-percolating to percolating or vice versa. This is often regarded as a phase change. Using the phase change in urban design and planning we can have cities which are more efficient because at this situation it formed a hierarchy of centers which help the city be more coherent and so it can work better. REFERENCES [1] Alexander, C. 1991. The new theory of Urban Design . Tash. Mashhad . [2] Bahraini, H. 1999.Urban Design Process. University of Tehran press . Tehran. [3] Lynch, K. 2005. A theory of Good City Form . University of Tehran press . Tehran. [4] Morris, j. 1989. History of City Form . Science and Industry University press. Tehran. [5] Ostrotovsky, V. 1999. Contemporary Urbanism . Centre for academic publication . Tehran. [6] www.calresco.org. ISBN: 978-960-6766-87-9 104 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 A Modified Arc Routing Problem for Highway Feature Inspection Considering Work-Shift and Overtime Limit Constraints MANOJ K. JHA, FRANCIS UDENTA, SIMON CHACHA, GAUTHAM KARRI Department of Civil Engineering Morgan State University 1700 East Cold Spring Lane, Baltimore, MD 21251 USA manoj.jha@morgan.edu; ankie2u@yahoo.com; schacha10@yahoo.com; kgautham.anand@gmail.com http://www.eng.morgan.edu/~catier Abstract: - Transportation problems are often defined in terms of delay and efficiency. In order to keep pace with deterioration of various elements within the transportation system especially highway appurtenances, it is expedient to monitor them periodically to evaluate their conditions. Valuable information obtained from such exercises drive subsequent actions to repair or replace these elements and keep highway appurtenances functioning as expected. Various past research efforts concentrated on highway maintenance optimization for pavements and other highway structures like bridges and drains. The recognition of the essence of highway road features in the overall safety of the highway system has yielded recent limited research efforts on strategies for optimizing maintenance of highway appurtenances. The efforts however, lack extensive work in the area of Maintenance Inspections (MI) for highway appurtenances. In practice, highway maintenance inspections occur within work-hour shifts in many government agencies and are often left to the inspector to schedule and organize his route. The inspector must leave from and return to a particular location (the office or yard) at the end of the work-shift. The implication is that the inspector spends a good portion of the work-shift traveling. This causes a cumulative waste of time and resources, and a huge cost to the taxpayers. In this paper we introduce a modified Arc Routing approach to a cost minimization inspection problem. We consider work-shift and overtime constraints in developing an efficient shortest path for implementing inspection schedules within a specified location. A unique algorithm is developed for solving the optimization model. The consequent solution is applied to real location in Baltimore, Maryland. Finally, a direction for future works is presented. Key-Words: - Highway inspection, Arc routing, work-shift, overtime, optimization, shortest path algorithm. 1 Introduction Highway inspection as well as maintenance reconstruction and rehabilitation (MR&R) should be done economically. More often in a highway network there are numerous segments (divided, undivided, one-way, two-way) see figure 1a and 1b that need to be inspected at routine intervals before appropriate MR&R actions can be undertaken. This paper deals with the development of a mathematical formulation and solution algorithms to the highway feature inspection optimization while considering work shift and overtime limit constraints. In previously published works, work shift and overtime constraint were not comprehensively considered. Jha et al applied genetic algorithm in developing a decision support system to choose the best economical route for maintenance inspection. However, the return path Figure1. A Typical Highway Segment in Baltimore, Maryland ISBN: 978-960-6766-87-9 105 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 condition was not taken into account. This leaves a huge void in ensuring policy compliance when attempting to implement the model. This paper takes the return arc and travel time into account as well as possible overtime considerations in determining the shortest path. In practice highway maintenance inspections occur within specified work shifts (such as 8-hour work day) in highway agencies. It is often left to the inspector to schedule and organize the inspection route for a given work-shift. The inspectors must leave from and return to a particular location (office or yard) at the end of the work-shift. The implication is that the inspector spends a good portion of the workshift traveling. This causes a cumulative waste of time and resources, which translates to great costs to taxpayers. highway features. Since then, several contributions in the area of operations research by Carnahan (1988), Liu et al (1997), Seresen et al (2000), Sriram et al (2003), and Hegazy (2004) have provided reference analytical models that help solve common transportation problems. Jha et al (2004) made significant progress in the area maintenance inspection and scheduling for highway features. Jha’s work was expanded upon by Abdullah (2006) by considering work-shift constraint. Maji (2007) applied budget constraints to maintenance scheduling. This paper is an expansion of Abdullah’s (2006) work on routing maintenance inspection. We introduce return path condition as well as overtime constraint. 2. Problem Formulation 2.1 Current Practices Most urban cities and counties have somewhat similar practices in how they approach inspection of highway infrastructures. They provide a strict work shift and provide vehicles that must be returned at the end of the work shift. Their policies generally discourage overtime. For clarity of perspective we will discuss current policies and practices in Baltimore City, Maryland. Baltimore City runs 8-hour work-shifts. Inspectors work in four (4) sectors: north-east, north-west, south-east and south-west. The quadrants are divided along Charles Street from north to south and along Baltimore Street from east to west. Each section is overseen by an inspector. The inspectors effectively use 6 hour within the work-shift for inspections on highway infrastructures such as pavement, bridges, road signs, guardrails and luminaries. These include inspections resulting from routine complaints and periodic maintenance inspections. Given budgetary and funding constraints, urban cities such as Baltimore City must prioritize its maintenance activities. The City launched operation orange cone in 2007 to put the City in the path of achieving its road repair and reconstruction goals. The focus of routine maintenance inspections are on major arterials. Secondary roads are not considered for this paper. This paper provides a modified arc routing formulation based on work-shift and overtime constraints. A probable solution using the shortest path algorithm will provide an insight on tackling a real life situation. A hypothetical shortest path example, based on a real case scenario will be presented followed by conclusion, recommendation and direction for future work. Figure 2. A typical urban network 1.1 Literature Review Numerous research efforts in highway maintenance have concentrated on improvement of perpetual elements of the highway infrastructure. Golabi (1982) pioneered a shift in research focus to ISBN: 978-960-6766-87-9 106 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 2.2 Formulation Assumptions Routing highway maintenance inspections is similar to Chinese Postman Problem. In this paper, we account for the limits (capacity) on the arcs therefore formulate the inspection routing problem as Capacitated Arc Routing Problem (CARP). To simplify the formulation, we have modified the CARP problem with the following assumptions: (i) The inspection crew is paid for eight (8) hours regular time (ii) The daily route is chosen such that inspection is completed in eight (8) hours with minimal overtime (iii) Each arc can only be inspected once (iv) Each arc can be traveled multiple times A valid tour contains the shop node twice (v) (vi) A valid tour contains arcs that connect the shop ⎧1 if arc (i, j ) is traversed ⎪ y1ij == ⎨ or inspected ⎪0 Otherwise ⎩ ⎧1 if T > 8 ⎫ y 2ij == ⎨ ⎬ ⎩0 Otherwise ⎭ ⎫ ⎪ ⎬ ⎪ ⎭ Based on the above, the objective function for the shortest path problem can be formulated as follows: min ∑ r i , j∈A )c y x + ∑ ∑ (T − t − t )c y x r i , j∈A r =1 1ij ∑ (t + t 2ij r r 1ij 1ij r ij r r =1 1ij 2 ij 2 ij 2 ij ij (1) Subject to, 2.3 Formulation Let Y= (V, A) be a directed graph with n+1 nodes of a network to be inspected (including the starting node (node 1) where the crew starts its inspection from). Each arc (i, j)∈A represents a segment between two intersections. For clarity a forward arc (i, j) represents movement from origin (office or yard) while backwards arc (j,i) represents movement back to origin. Within a work-shift, T the time required to traverse a given arc (i,j) is tij, while the time required to complete the necessary inspections along the arc is cij. Figure 3 A typical network showing a shop node. 7 8 7 8 10 k∈N ∑x r r ki − ∑ xik = 0, ∀i ∈ N , r = 1,2,..., r ( 2) r k∈N ∑y v =1 r 1ij ,∑ y =1 r r =1 2 ij r ∀(i, j ) ∈ A ( 4) y1ij = {0,1} ∀(i, j ) ∈ A (5) y 2ij = {0,1} T > (t1ij + t 2ij ) ∀(i, j ) ∈ A (6) T,C1ij,C2ij, t1ij, t2ij, y1ij, y2ij, xij > 0 Non-negativity constraint (7) Where; A: Set of arcs V: Set of nodes C1ij:Hourly cost in regular time ($/hr) C2ij: Hourly cost in over time ($/hr) t1ij : Time to traverse arc (i,j) t2ij : Time to inspect arc (i,j) T: Cumulative work-shift duration y1ij: 1 if arc (i,j) is traverse or inspected, 0 otherwise at regular time y2ij: 1 if arc (i,j) is traverse or 1 Shop 9 1 6 6 4 2 5 3 2 3 5 4 ISBN: 978-960-6766-87-9 107 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 inspected, 0 otherwise at overtime xij : Number of times arc (i,j) ∈ A is traversed v : Number of routes to be traversed The objective function expressed in equation (1) minimizes total travel time required to traverse an arc and complete the necessary inspections. Equation (2) and (3) define time needed to traverse an arc and the time used to complete inspection along the arc respectively. Equation (4) guarantees that inspection occurs along each arc at least once. Equation (5) states that an arc must be traversed and inspected during work-shift, T and the overtime constraint is given in equation (6). Equation (7) is the non-negativity constraint. 3 Problem Solution combining the formulation assumption and the constraints. The steps for the program for the solution model are as follows: Step 0 Give an integer between 1 and n, n1 Step 1 If n1 < 3, then go to Step 0 Step 2 Random generate n1 nodes Step 3 Check if tour is valid otherwise go to Step 2 Figure 4 shows a typical highway network in Maryland. We can apply the resulting model in establishing to most efficient way to inspect the network at minimal cost while taking into account work-shift and overtime constraints. Figure 4. A Typical Highway Network in Charles County, Maryland The general solution approach is heuristic. We consider real life conditions at specific location in Baltimore, Maryland. Solution approach is two-tier: first we use Floyd-Warshall method to determine the shortest path, feasible network and optimal values for the travel time and inspection time. The values shown in Table 1 are the optimal total time for the network shown in Figure 3. Table 1 Optimal Values of Total Time Arc Travel Time Inspection Total Time (min) Time (min) (min) 1 25 45 70 2 15 35 50 3 10 25 35 4 20 30 50 5 30 45 75 6 17 35 52 7 25 45 70 8 35 50 85 9 25 45 70 10 15 35 50 The above optimal values can be plugged into to objective function to obtain the optimal cost for efficiently covering all the arcs in a timely manner. For more complex networks, we can develop a solution model using Genetic Algorithm by 4 Conclusion Considering the effect of the attributes of a shop node within a highway network and constraints imposed by the work-shift and overtime, highway maintenance inspections can be routed in a cost efficient manner. Optimization of time in performing inspection activities translates to direct optimization of resources and drastic reduction of waste. However other aspects of planning such as scheduling must be in place for the true impact of this research to be fully appreciated. Consequently, we recommend further research in highway maintenance inspection scheduling. ISBN: 978-960-6766-87-9 108 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 References: [1] Carnahan, J.V. (1988). Analytical Framework for Optimizing Pavement Maintenance. ASCE Journal of Transportation Engineering, 114 (3), 307-322.State the benefits to your audience for taking this action [2] Golabi, K., Kulkarni, R.B., and Way, G.B. (1982). A Statewide Pavement Management System, Interfaces, Vol. 12(6), 5-21 [3] Golabi, K., and Shepard, R. Pontis: A System for Maintenance Optimization and Improvement of US Bridge Networks. Interfaces, Vol. 27, pp. 71-88, 1997 [4] Jha, M.K., and Abdullah, J. A Markovian Approach for Optimizing Highway Life-Cycle with Genetic Algorithms by Considering Maintenance of Roadside Appurtenances. Journal of The Franklin Institute, Vol. 343, pp. 404-419, 2006. [5] Jha, M.K., Abdullah, J., and Dave, D., GIS Application in Developing a Roadway Feature Inventory Program. Proceedings of the 2004 ESRI International User Conference, Paper Number 2128, San Diego, CA, 2004b. [6] Haghani, A., Shafahi, Y. (2002). Bus Maintenance Systems and Maintenance Scheduling: Model Formulations and Solutions. Transportation Research Part A: Policy and Practice, 36 (5), 453482 [7] Hall, R.W. (2000). Scheduling and Facility Design for Transit Railcar Maintenance. Transportation Research Part A: Policy and Practice, 34 (2), 67-84 [8] Hegazy, T., Elhakeem, A., Elbeltagi, E. (2004). Distributed Scheduling Model for Infrastructure Networks. Journal of Construction Engineering and Management, 130 (2), 160-167. [9] Liu, C., Hammad, A., Itoh, Y. (1997). Maintenance Strategy Optimization of Bridge Decks Using Genetic Algorithm. ASCE Journal of Transportation Engineering, 123 (2), 91-100 [10] Maji, A and Jha, M K (2007) “Modeling Highway Infrastructural Maintenance Schedule with Budget Constraint”, Transportation Research Record 1991, Journal of the Transportation Research Board, 19-26 [11] Srensen, D.J., Faber, M.H. (2000). Reliability – Based Optimal Planning of Maintenance and Inspection. Proceedings of the U.S. – Japan Workshop on Life-Cycle Cost Analysis and Design of Civil Infrastructure System, Honolulu, Hawaii. [12]http://www.aaroads.com/midatlantic/maryland999/moravia_sb_at_us040_wb.jpg ISBN: 978-960-6766-87-9 109 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 Reconfiguring the Existing US Road System for Sustainable Living DIANE JONES, RLA MANOJ K. JHA, Ph.D., P.E. Department of Civil Engineering Morgan State University 1700 East Coldspring Lane Baltimore, Maryland 21251 USA Diane.jones@morgan.edu; manoj.jha@morgan.edu Abstract:-This paper will explore a model linking compact development and mass transit to enhance sustainable living. The current land development model is an unsustainable one, built on a continuous outward expansion, and the ever increasing need for transportation capacity in the form of roads and highways. This trend threatens quality-of-life, including economic and environmental health. Key-Words: - New Urbanism, transit-oriented development, sustainability, accessibility, highway systems, mass transit the release of carbon monoxide and other harmful gases. Cumulative emissions cause 1. Introduction tremendous environmental damage, The interstate system has been the most including climate change, damage to water important influence on the US metropolis in bodies, crop damage, and habitat the last 50 years (Lewis 1997). The mass destruction. Coincidently, auto induced transit system was neglected in favor of sprawl causes the loss of wetlands, open expanded highway development. The space and valuable farmland. Continuing to government paid 90% of the cost of this develop communities around and dependent expansion. Instead of saving the central upon auto transportation is not cities as proposed, the highway system environmentally or financially feasible. enhanced the growth of suburbia. The move of population and employment from the 2. Problem Formulation urban core to outer urban and suburban Interurban and outer urban movements areas increased auto usage which, in turn affect the spatial patterns, structure and rendered the roadway system unable to development of cities and their outlying efficiently handle the increasing number of areas. Mass transit becomes less economical cars. given the current development pattern. The economic and environmental Changing the way in which communities are cost of expanding roads to keep up with developed will change the way congestion is an ever expanding problem. The cost of fuel is continually rising and transportation is designed. Building a may soon become prohibitive for most network of mass transit (buses, streetcars, Americans. Additionally, automobiles have electric trains, etc.), that interconnect a negative effect on the environment through development and follow existing road ISBN: 978-960-6766-87-9 110 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 patterns, can create economically and environmentally sustainable communities. A model can be developed which builds a passenger rail network, based on two principals: • Using existing highway and road systems to construct rail. • Using “new urbanism” as guide for land development. The US highway network consists of 4 million miles of roadways and streets. The interstate system consists of only 1% of all highway mileage, but carries 25% of all total vehicle miles of travel (Lewis 1997). Much of the system is currently in need of upgrading and repair. The system could be retrofitted to handle rail. Funds proposed for new major road expansion and construction could be shifted to infrastructure upgrades to accommodate mass transit systems. Automobiles have come to dominate personnel travel in the US. Travel by personal vehicles, including journeys to work and inter urban travel, make up nearly 80% of trips. Rail and transit together account for less than 1% of US travel (Hanson, 2004). In contrast, European cities have high population densities and well developed rail systems. American’s have clung to cars because current configurations don’t offer the flexibility or convenience of routes and times that are desired by travelers. Using the existing highway system for new mass transit and rail lines would offer the desired flexibility and convenience. Rail systems, can also bolster consumer budgets with decreased costs for gas, and automobile maintenance. Transitioning away from automobiles and increasing mass transit can be achieved by changing the ways communities are developed. The use of “new urbanism” planning methods to decrease sprawl and halt the development of new roads can help in this transition. “New Urbanism” as defined by the Congress is characterized as the reordering of the built environment to create complete cities. It includes infilling cities and suburban areas to create compact development that offers easy access. Compact development promotes walking, biking, and the use of mass transit, the keys to sustainable communities. Therefore, existing and new communities must be configured to a scale and density to facilitate a new transit model. The revitalization and densification of existing cities and suburban development focused on mass transit as a means to travel the same distances and routes as cars can be achieved through this reconfiguration. The following problems must be addressed in formulating a new model for transit based development: • How to share the existing road system with mass transit? • How can these new “inter-modal” systems be engineered? • How to create a hierarchy of transit modes that interconnect? • How to create infill development that provides easy access, collection and distribution from transit nodes? 3 Problem Solution A typical outer urban neighborhood in the US (Fig. 1) can be used as a case study to apply a network of transit on existing roads, including bus streetcar, light rail. This system of transit in combination with mixed use and infill development can create a more livable, environmental and economically sustainable community. Similar systems currently existing in other parts of the world, especially European cities, are excellent examples. ISBN: 978-960-6766-87-9 111 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 Fig. 1 Baltimore Maryland 3.1 Design Considerations for Retrofitting Roadways to mass transit. Depending on the functional differences, a wide variety of technical standards must be used for the various modes (streetcar, bus, light rail, etc) comprising a mass transit system. Existing roads must be redesigned with transit vehicles in mind. Lane widths, turning radii and pavement thickness should reflect the type of transit vehicle that will use the street. A variety of design considerations must be taken into account when developing a new mass transit system based on compact development. These include the following: • Traffic and safety • Functional classifications of the roadway • Projected traffic volume and composition • Required design speed • Land development patterns of surrounding community • Multimodal needs of surrounding community • Topography of the surrounding land • Capital costs for construction • Agency funding mechanisms • Public involvement, review, and comment • Environmental considerations • Right of way impacts and costs Design features of a transit system influence visible features and directly affect capacity and operations, social acceptance to users, owners of adjacent land and the general public. Pedestrian activity, access to and interaction with these systems must be designed and criteria and standards set. Design issues associated with mass transit that relate to the locations where riders get on and off the transit vehicles must be considered. Several strategies can be used to incorporate and give preference to transit vehicles. The specific strategies applied will depend on the context, the amount of road space available, pedestrian flow characteristics, and access to adjacent parcels of land. Transit locations will depend on factors such as physical design of the intersections, and the proximity of passengers. Adequate passenger access by walkways must be provided for. 3.1.1 Rail Use on Street and Highway Rightof-ways Light rail transit consist of rail vehicles that can operate on exclusive rights-or way at ground level, on aerial structures, in tunnels, or on the street with other vehicles. Because these conditions currently exist it would not be prohibitive to have light rail operate on existing roadways. Roadways can be adapted to transit only use and also to mixed traffic situations which allow automobile use to continue. The design of light rail into roadways should consider the safety of passengers as they board and disembark. The directional movement of automobiles and the light rail vehicles must also be considered, especially when there is a mix of traffic including motor vehicles and mass transit. The vehicle alignment should be designed depending on whether the rail system is on a local street, arterial, collector or highway. ISBN: 978-960-6766-87-9 112 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 Figure 2 Figure 6 Figure 3 Figure 7 3.1.2 Bus ways on Streets and Highways Figure 4 The railway could be designed with complete separation, limited motor vehicle and pedestrian crossings, and alignments that allow for mixed-flow operations with automobiles and pedestrians. A bus way is a special roadway designed for the use of buses. Bus ways often have there own right of ways with grade separations and online stops. They mimic the rail service in many ways and link to and often are the first stage in building a rail service. Bus ways typically have two-way operation. Flow of operation, access to stops and platforms must be considered in design. 3.2 Design Standards for Compact and Transit Oriented Development Neighborhoods should be compact and detailed to encourage pedestrian and mass transit activity. Streets should be laid out in a grid pattern forming blocks where entrances front the street rather than parking lots. This pattern will allow transit to connect neighborhoods and city districts to each other and the surrounding area. Compact development is also characterized by a mix of land uses. A diverse use mix should occur, including a Figure 5 ISBN: 978-960-6766-87-9 113 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 range of housing types. This mix of uses creates diversity. Diversity is the key to the success of a city district (Brook, 1961). It brings people into the area for varying purposes and contributes to the fabric and vibrancy of a neighborhood. This diversity and proximity of uses are further enhanced by creating a physical link. This could be a pedestrian way or promenade or a transit line, to connect these uses. A public square, civic building or a transit hub should become the center of this development. A mass transit system, unlike an interstate system, often aids in the creation and preservation of public space. Decreasing traffic lanes and widening sidewalks can recapture public space. Tangibly, the widening of the side walk would allow for planting of trees, additional lighting, benches, and other site amenities that enhance travel and create a since of place. This would in affect increase pedestrian activity and enhance the environment along the transit corridor by making it active, safer and visually pleasing. Fig. 9 Design considerations for transit oriented and integrated development include the following. • Infill vacant lots with structures to create continuity of the pedestrian experience. • Balance on street parking with other uses • Change traffic signal timing to improve mass transit and pedestrian access. • Improve use of parking through changes in enforcement or regulation • Widen sidewalks or provide sidewalk extensions at crosswalks, better balancing pedestrian uses with other uses (vehicles, transit vehicles, bicycles, deliveries, etc.) • Increase security by providing more uses for and activities, which will increase the number of people present. • Make accommodations for bicycle users (bike lanes, lockers, storage rack, etc). Fig. 8 Mixed use hubs or areas with housing, commercial and transit located in a mile radius would also lessen automobile use. ISBN: 978-960-6766-87-9 114 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 4. Conclusion The key to refiguring the US road system for sustainable living is through instituting rail lines and other forms of mass transit, and infilling existing communities. Infill and compact development enables a community’s growth to be channeled into a form that is compatible with existing urban neighborhoods, encourages mass transit, and is economically and environmentally feasible. Infilling, existing outer urban and suburban communities not only controls sprawl, but is cost effective. There is less per capita spent on infrastructure and utilities than on the typical sprawl due to the compact nature of infill. There is also an increased tax base due to increased density. The increased tax base and density makes providing mass transit more feasible. Mass transit, as previously discussed, offers enumerable benefits including cost savings in energy resources, and road and parking expansion. Compact transit oriented communities have less traffic congestion, are more walk-able, and have a greater sense of security and place. (Jane Jacobs, The Life and Death of Great American Cities). References: (1) “BURHA Report on the Expressways Consultants’ Preliminary Engineering Report for Interstate Highways 70N and 95,” Baltimore Urban Renewal and Housing Agency. Baltimore, Maryland. January 12, 1962 (2) Brooks, Jane. The Death and Life of Great American Cities. New York: Random House, Inc. 1961 (3) The Congress for New Urbanism.2008. “New Urban Living” http://www.newurbancommunities.com/ newurban/ (4) Rachel F, AICP and Chimere LesaneMathews, “West Baltimore and Transit Centered Community Development: A Review of Community Plans and Exploration Opportunities” (prepared for the Transit-Centered Community Development Initiative of the Baltimore Neighborhood Collaborative, March 2006). (5) Francis, Mark. Case Study in Land and Community Design Urban Open Space Designing for User Needs. Washington: Island Press, 2003 (6) Graham, Stephen and Simon Marvin. Splintering Urbanism: networked infrastructures, technological mobilities and urban condition. New York Rutledge: 2001. (7) Hanson, Susan and Genevieve Giuliano. The Geography of Urban Transportation. Third Edition. New York: The Guildford Press, 2004. (8) Lewis, Tom. Divided Highways: Building the Interstate Highways, Transforming American Life. New York: The Penguin Group, 1997. (9) McNichol, Dan. The Roads That Built America: The Incredible story of the U.S Interstate System. New York: Sterling Publishing Co. Inc., 2005. (10) NewUrbanism.org. 2008. “New Urbanism” http://www.newurbanism.org/Fram496683-sustainabilitypage496683.html (11) Reid, Robert L, “Paving America From Coast to Coast,” Civil Engineering Magazine, June 2006. (12) Seely, Bruce Edsall. Building The American Highway System: Engineers As Policy Makers (Technology and Urban Growth). Pennsylvania: Temple University Press, 1987. (13) Wright, Paul H and Dixon, Karen K. Highway Engineering. New Jersey: John Wiley and Sons, Inc. 2004 ISBN: 978-960-6766-87-9 115 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 Adaptation of New Dwelling Houses in East Towns to the Historical Environment (Case of Iran and Azerbaijan Republic) JAVAD ABDOLHOSEYNI Department of Architecture Azerbaijan Architecture and Construction University Department of Architecture, Azerbaijan Architecture and Construction University, Baku AZERBAIJAN javad_abdolhoseyni@yahoo.com Abstract: -Evolution process of living places depended on modification temp of social-economical structure of the society on the history. Cities obtained the most stable structure with changing for the some reasons (economical potential, foreign relations, natural-climate specification and etc.) and determined its types within one territory. On basis of this the main development stages of Iran and Azerbaijan historical cities and formalization conditions of living buildings of Moslem cities, including development problems of living architecture of the historical cities are researched in formalization process and structural specification of Moslem countries cities. So that the main modification occurred in architectural and town-planning profession under the dissemination of Islam at the beginning of 7th century and brought innovations to the social living life of the cities. Then improving ways are researched in types of living buildings of Iran and Azerbaijan historical cities, in volume-place and architectural specifications, including volume-place of living buildings of two historical cities (Baku, Tabriz) under building practice and improvement ways of historical cities in 19th. -20th centuries. While development stages of researched cities are followed, it becomes clear that though type and architectural style of their living buildings are difference, it is possible to divide the living houses into four types: a) Living houses, belonging to people architectural models; b) Individual large and handsome buildings; c) low and medium storied living houses, formalizing the city streets and giving in row form from the end of 19th century; d) High-storied living houses, sometimes playing dominant role. Thereby, Application opportunities of new ideas and adaptation principles of new living building to existing condition are researched in reconstruction of resolution principles of new living buildings in historical Moslem cities. Usage of social-cultural and new technological opportunities of their chosen cities and their “quality modification”, appointed under stable method, should be prescribed mainly. As well as progressive practice (“ideal city” idea), heritage and main traditional specifications of Iran and Azerbaijan cities should be considered here. Key-Words: - Dwelling House, Historical Environment, Adaptation, Urban design, Traditional architecture 1 Introduction Throughout the history, the process of populated area’s evolution has been dependent upon rates of changes in the social and economic formations. In the course of historical evolution the social and economic formations replaced each other to influence mode of life and social structure of the population, as well as architectural treatment and spatial-volumetric lay-put of the architecture. Since the 7th century, owing to the spreading of Islam in these countries, significant changes took place in the architecture and townplanning, scores of philosophically-colored novelties intruded into the social life of towns. This, in turn, gave impetus to the construction of “quarter” system of townplanning, the construction of religious erections and big palaces, manors. [9] (Fig. 1) Owing to the fact that many Iranian and Azerbaijani towns were laid along the international caravan routes, ample opportunities for the development of trade, handicrafts and architecture were available. Some of the towns, thanks to their economic-political position, were granted a status of capital to thus consolidate their threestage structure as fortress-towns (internal fortress, internal city and external City-Rabat). [7] Today, it is essential to carry on studies in identifying up-to-day, progressive methods of construction to comply with growing social needs of the population. Thus, newly erected buildings are not consistent with traditional way of life and flats with all conveniences. Hence, there is a great need in improving architectural treatment and taking adaptation measures. The construction of dwelling houses is a large-scale problem; a research work has been done by a good many researchers to explore spatial-volumetric aspects of dwelling houses, climatic conditions and the role of building materials in the process. However, these materials evaded such an important and urgent question as conformity of new dwelling houses of historical ISBN: 978-960-6766-87-9 116 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 towns. Meanwhile, it was residential buildings that helped shape functional life and architectural image of towns. With the purpose of solving these problems, it is essential to strengthen renovation processes in the towns of Iran and Azerbaijan (Tabriz, Bushehr, Yazd, Baku, Gyandja and Sheki) and uncover new trends in the development of housing construction. everyday traditions, as well as architectural and townplanning models of these towns that make it possible to compare them and establish a fact of their common path of development. Note that these towns keep on developing to comply with advanced architectural concepts and special emphasis on insolation-aeration principles; flats with all conveniences, etc. Besides, a positive experience of the Iranian folk architectural finds its parallel in the historical quarters of Baku. [3] Fig.1: The structure of Baku Castle (Kh.Truzson, 1796) A key question here is to identify adaptation potentialities of new multi-storied buildings in the said towns of Iran and the Azerbaijan Republic to historical environment, to identify new methods of effective use of dwelling houses. It is essential to study the experience of bringing new multi-storied residential buildings into conformity with population’s mode of life, reveal new adaptation methods based on proper assessment of adaptation trends from spatial-volumetric and artisticaesthetical points of view and finally develop scientifically grounded and practically approved recommendations. For this to happen, it would be appropriate to explore archival documents, materials of building organizations and photo-fixation studies of populated areas of the above towns. The scientific-practical novelty is that some regularities of appropriate adaptation methods to bring new residential buildings into conformity with existing surroundings have been identified, and adequate design models adjusting for new technological opportunities and traditional architectural principles, have been offered. An important question is the necessity of carrying out a comparative analysis of historical towns of Iran and Azerbaijan. (Fig. 2 and 3) Fig.3: An alley in traditional urban texture of Baku (19th Century) 2 Adaptation analysis in Iran and Azerbaijan It would be appropriate to consider issues as follows: classification of types of dwelling houses; identification of improvement principles; modern architectural styles of dwelling houses; composition-construction and artistic-aesthetic design principles. An analysis of major stages of the development of historical towns of Iran and Azerbaijan is illustrative that their historically established mutual relations had a great influence on the art of architecture in the region to give impetus to the improvement of building conceptions on subsequent stages (7th century), i.e. origin and dissemination of the Islamic religion. The process that started with laying down the foundations of towns and fortresses, including the construction of palaces and mosques; erection of fortification works on rising ground with walls and doors looking four directions – all these had much in common with other principal towns of the country.A distinctive feature of dwelling houses of Moslem towns were their independent lay-out; organized character and, most important, 1-2 story buildings with inner courtyards and inwardly turned houses which harmoniously blended with surroundings. [7] The grouping of dwelling houses has its effect on insolation conditions when adjusted for three main factors: comfortableness of the dwelling house, functional and artistic-aesthetic design of rooms; harmony between surroundings and the dwelling house. Up to the end of the 18 century, towns of these two neighboring countries preserved the quarter system which accounted for the special structure and the related Fig.2: Traditional city districts of Yazd-Iran It is similarity of climatic conditions, religious and ISBN: 978-960-6766-87-9 117 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 character of outward appearance of traditional buildings. However, in the 19-20 century architectures of Iran and Azerbaijan came to vary and assume original features. Thus, Baku being based on principles of the development of Moslem towns notwithstanding, following the middle ages and in connection with oil boom, it began imbibing architectural styles of the European countries. [5] The first stage of Baku development covers a period from the 11 to the mid-13 centuries. Walls of the city and the Maiden’s Tower were reconstructed; foundations of the palace complex laid down and the inner fortress arose around it. The same period saw the construction of the city’s oldest and first minaret – Synyg-kala (1089); building of a water pipeline and a Bayil Kasri (1234) on an island in the bay. One can say with certainty that the main street network and spatial-volumetric structure of Baku formed in the 15 century. The fact that Tabriz performed major cultural, commercial and scientific functions in the Moslem East found its parallel in the lay-out of the city. Note that architectures of Iranian towns go back to their traditional sources with a special emphasis on “quarter” structure and courtyard space. Thus, softening of courtyard climate is achieved due to open pools and fountains. Besides, vertical pits were drilled inside buildings of Yazd and Bushehr towns to thus regulate air climate. Still influencing residential environment of Iran remains to be everyday life and religious factors. These buildings are outwardly closed and inwardly open for family members. Beyond any doubts, the Iranian architecture laid a special emphasis on adequate direction of wind and Mecca (Qibla), regularity of habitation, arrangement of rooms in the courtyard and their functional purposes, i.e. aspiration to conserve a traditional mode of life. A distinctive feature of the Iranian and Azerbaijani traditional architecture is the closed character of spatialvolumetric structure of dwelling houses and rich design inside the courtyard space. Factors directly influencing the formation of dwelling houses are as follows: tradition, local relief and social economic conditions. These determine architectural images and artisticaesthetic features of dwelling houses. There are numerous examples of medieval architecture in the town-planning and compositional structure, as well as the interior of residential architectures of modern Iranian towns. Since the 20 century, the architecture based on the latest social concepts does broaden its economic potential and assume new straits at the expense of the construction of administrative buildings. Thus, depending upon climatic and social conditions, dwelling houses are treated to comply with volume and space requirements. Proceeding from local building materials, these houses are reflective of architectural forms based on folk life, everyday life and limited requirements; adequate use of local building materials; type of construction in accordance with appropriate traditions and design tending toward inner space. The most typical features of the Iranian housing architecture are rectangular gardens and decorated pools located in the courtyards. A special principle was laid in the use of 2-, 3-, 4-, and 5-door rooms across the courtyard. [6] (Fig. 4) Fig.4: Interior plan of the rooms of different type 3 Comparison of Housing Construction Features in Iran and Azerbaijan Housing construction features of the Iranian historical towns are currently being studied; they are examined from present-day point of view. Spatial-volumetric distinctions of these towns are that the local population, in an attempt to adapt them to their own mode of life, looked for effective methods to attain this goal. A typological classification of dwelling construction and buildings of the Azerbaijani historical towns is being carried on to uncover their architectural features. Note that methods of construction typical for Tabriz in the Middle Ages are still maintained, because there are numerous facts which bear out their compliance with today’s social needs of the local population. A special emphasis is laid in Tabriz on apartments being shaped as the inner space. The question is that the resolution of household problems within the framework of flats is a matter of paramount social importance one of the constructive features of flats is the creation of splendid architectural environment at the entrance, particularly, location of the hall higher than the ground floor, covering of a ceiling with convex glass and presentation of the nature in the hall part of the flat. Organization of favorable micro-climatic conditions in the city of Tabriz is frequently dependent upon architectural design and interaction of elements in private houses. For this reason, not only houses but also courtyard, porch and even transitional forms may be presented in the form of block-model type. It is essential to take into account the rational room arrangement, adequate kitchen location, special design of summer areas and other factors. In this respect, the traditional ISBN: 978-960-6766-87-9 118 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 Iranian experience may as well be applied in our days. [7] The analysis shows that together with functional and spiritual components within the adaptation process of the Tabriz historical environment, an appropriate account has to be taken of regularities of its local layout structure, transport, services and vegetation systems. When analyzing the problem, it is essential to put forward some scientifically grounded proposals, including assessment of ventilation and insolation factors; provision with summer rooms in-between flats of blocked construction; two sectors for big families with separate exits. A positive feature of the Tabriz experience is that the new residential buildings are located slightly far away from historical centers. Common features of houses in Yazd and Bushehr are as follows: physical and visual interrelation between houses and environment; girder casting system of the building; effective use of “Tarma” (type of porch); “Shenashil” (L-shaped balconies on the juncture of facades); “Balakhany” (summer rest-room above a roof of the building); “bun” (rest place on a roof of the building); “Pishbun” (open rest place slightly above a roof of the building) and “Serpelle” (an attic storey on a roof). And finally, ventilation holes of different forms in four angles of the house. Depending on the number of rooms, inner and outside courtyards are differently designed: buildings are located in 1, 2, 3, 4 parts of the courtyard. [10] (Fig. 5) Fig.5: Forming of design and structure at dwelling house Types and architectural features of residential buildings in the historical towns of Azerbaijan are that despite the presence of medieval irregular system, subsequently there were used local “irrationalism” and “rationalism” being imported from abroad; construction methods contributed to the harmoniousness in the urban environment. At that time, European “baroque”, “modern”, “gothic” styles were used in some dwelling houses and administrative buildings. In the subsequent period of the Baku architecture both town planning or administrative buildings found their parallel in some aspects of the urban construction. [2] An eloquent testimony to the fact are components of residential buildings which imbibed artistic features of the capitalist epoch of the late 19-early 20 centuries of Baku as well as local and European styles (“chatmataglar”, “portal”, “tagband”, “gunbez”, various types of ornament). However, all these showed their positive influence till the 1920s only. The point is that later on one-sided development of urban industrialization violated this progressive process and negative changes manifested themselves in the town planning and inner spatial structure of dwelling houses. Thus, one type dwelling construction and complexes were widely spread in the urban environment under destructive socialism standards and the Azerbaijani architecture was brought into conformity with the Soviet ideology to ignore the glorious architectural history. In the early 20 century, after the revolutionary changes in Azerbaijan the historical towns of the country faced with numerous problems. Enlargement of towns and rise in the number of population aggravated the provision of urban residents with housing. An evidence of this was the overuse of one-type construction and new buildings’ disagreement with mode of life of the population and finally the loss of individuality of these buildings. In the early 20 century, a new style titled “modern” and reflecting improvement of the construction came to be spread in Baku. This situation was typical not only for Baku, but also for the towns of Gyandja and Sheki. Following the gaining of independence in the 1990s, against the background of general economic development there had widely been spread large-scale reconstruction measures and various projects and constructions were put forward. [5] In relying on major factors in the design of inner courtyards of buildings, it is possible to identify adaptation potentialities of new buildings to comply with historical towns of Baku and Tabriz. From this point of view, with the purpose of uncovering factors which play a decisive role in the everyday life, rehabilitation and labor processes, it is utterly crucial to acquaint with actual reconstruction ideas. For this reason, reconstruction measures should be based on new social adaptation methods and forms with special emphasis on intensive change methods. However, positive and negative aspects of the populated areas having been built in the 1930-80s are so largescale that one cannot ignore them. As a result of reconstruction measures that started in the second half of the 20 century, large-scale construction is underway in Tabriz. With the object of preventing city’s enlarging westwards the mixed storey order is applied to the newly built complexes. It is essential to give preference to individuality, inner space rationalism and mobile constructions. It would be appropriate to make use of the latest western technologies. (Fig. 6) ISBN: 978-960-6766-87-9 119 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 Fig.6: Climatic dwelling house in Baku city (author: V.Sulimova The analysis is indicative that it would be helpful to pay greater attention to aesthetic quality of outward appearance of the apartments being built, so it is comforting that Yazd and Bushehr towns with their specific climatic and economic conditions take this circumstance into consideration. [1] To remove some contradictions on this track, it would be appropriate to take into account population’s social and everyday needs and make necessary changes, particularly, select a method aimed at identifying advantages of adaptation measures. When adjusted for Baku conditions, it is necessary to maintain tendencies toward the use of stepped structure in the construction, placing of communication services in-between floors, increase of the number of special subsidiary rooms and provision a space between entrances with summer rooms. [3] Account has also to be taken of questions arising from the use of ventilation an insolation in the dwelling houses of Tabriz. The point is about a space with special interior due to the use of water in the traditional houses in summer-time and location of kitchen on the ground floor, or in two places in present-day buildings. Compliance with modern town-planning rules, provision with three-branch space and, finally, more effective use of artistic architectural components – all these are designed to create comfortable conditions and necessary conveniences for the population. [1] With the purpose of preventing intensive enlargement of big towns due to the use of new concepts in the reconstruction of historical towns and protecting historical monuments, the reconstruction processes should be based on application of “centrifugal” methods. In general, a special stress should be laid on architectural elements which play a decisive role in ensuring stability of residential type buildings. Thus, placing of these largely functionally invariable elements inside residential buildings is sure to be reflected in interior and exterior. Allowing for social requirements, a purpose of raising the quality of buildings has, to a degree, been attained commencing from the late 20 century. As a whole, positive tendencies currently manifest themselves in the environmental organization. The point is about refusal from closed character of dwelling through the use of 3-spatial concept and transition to the open type system; more effective use of elements to raise the quality of comfortableness; apply more rational treatment of in-between rooms’ relations; take space and volume-adjusted measures. With that end in view, it would be expedient to apply methods of experimental designing in the restructuring process. [2] At present, there is need in unfolding more flexible and goal-oriented activities to bring new residential buildings of the historical towns of Azerbaijan into conformity with present environment. For more effective use of zones in the environs of Baku with heightened relief difficulties it is essential to take into account natural geographical factors when erecting new dwelling houses and allow for some corrections in the orientation of new buildings and 3-spatial treatment. (Fig.7) Fig.7: Modern Modern appearance of traditional elements In the Iranian historical towns, as distinct from this, more widely used in the houses designed for one family only is the independent “quartering” of flats. Thus, it would more appropriate to construct terrace-type buildings in Bushehr in the new districts, and improve dwelling houses to comply with traditional methods in the historical center of the city. It would also be expedient to carry out construction operations in big towns on the basis of mixed typological principle. Adaptation of new dwelling houses to the historical environment would yield rather positive results.[4] The construction of one-storeyed dwelling houses in the town of Yazd in pairs is a progressive tradition of the architectural heritage to assume a particular importance under hot climatic conditions. For instance, when microclimatic conditions and traditions are allowed for adequate of design-space organization, a building looks attractive from artistic-aesthetic point of view. For towns with planned development, the best solution would be the construction of buildings in strict order. However, allowing for different designs and materials, adaptation methods should vary as well. Sometimes, it is suffice to improve some elements to thus ensure required comfortableness. ISBN: 978-960-6766-87-9 120 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 Hence, it is necessary to apply a method of differentiation in the course of restructuring and make a selection of new residential buildings in several directions. In this respect, it would be appropriate apply both radical and locally-adjusted reconstruction methods. [8] One of the ways to remedy the situation is to draw up principles of reconstruction specially designed for centers of towns and outlying regions. For instance, several housing models are offered for central areas: private houses as major construction to comply with new construction standards; dwelling houses with free courtyard areas inconsistent with respective standards; dwelling houses with improved design contrary to new construction standards; mediumstoreyed quartered residential structures; high-storeyed one-quartered residential structures.[4] (Fig. 8) erecting its new dwelling houses to thus meet its social requirements. Meanwhile, there is a great need in exploiting progressive traditions and rich elements of the architectural past. 3- When retracing the development stages of the historical towns of Azerbaijan – Baku, Gyandja and Sheki, it becomes obvious that despite differences in their architectural image and styles the dwelling houses may be classified into four types: 1. dwelling houses pertaining to the folk architectural patterns; 2. individual estate-type buildings; 3. low and medium-storeyed rowshaped dwelling houses which form streets of the towns; 4. sometimes dominant, high-storeyed dwelling houses. 4- When retracing the said types from adaptation to the historical environment standpoint as exemplified by Baku, one can single out mainly four typical stages where they became apparent: 1. adaptation traditions to the environment of pre- and during the 19 century of Baku; 2. adaptation tendencies of Baku in the late19early 20 centuries on the basis of romanticism and European architectural motifs; 3. indifference to adaptation processes in the large-scale housing construction of Baku on the former Soviet space; 4. wide adaptation measures in the dwelling houses built on the basis of new technologies and materials. Fig.8: Compactness of high-rise buildings losing expressiveness of traditional and religious monuments In considering the above-mentioned, all three construction types are offered for Tabriz in the future; Malikane (estate) type houses are offered for Yazd; high or mainly low-storeyed houses are offered for Bushehr; lofty or medium-storeyed quartered buildings are offered for Baku. 4 Conclusion and recommendations 1- Analysis of Tabriz architecture is indicative that no appreciable changes occurred in the outward appearance and internal design of dwelling houses built in the 19-20 centuries. Owing to the preservation of many principles of folk architecture, the adaptation process proceeded rather smoothly. However, since the mid-20 century housing construction was widely spread and at the same time a greater portion of the population gave its preference to individual dwelling houses. Today, against the background of historical environment there have been uncovered adaptation opportunities for new buildings. 2- Present-day dwelling houses in the Iranian towns of Yazd and Bushehr with their specific architectures are based on popular traditions. However, today the population is engaged in adaptation actions when References: [1] Abdolhoseyni J.E. Adaptation Characters of New Construction Types to Historical Environment, Journal of URBANIZM, No.10, 2007, pp. 117-118. [2] Afandizada R.M. Historical Region and New Buildings of Baku, Journal of Contemporary Democratic Collection, Materials of international conference. Baku: 2004, pp. 253-256. [3] Alizada G.М., Architecture Theory and Building Practise in Azerbaijan, Baku, Maarif Publishing House, 1986. [4] Etenko V.N. Many-Storied Living Houses in Building of Foreign Countries, Moscow, 1967. [5] Fatullayev Sh.S., Living Houses at Building of Azerbaijan Towns at the Beginning of XIX-XX Centures, Baku, 1963, pp. 118-120. [6] Ghezelbash, M.H., ABC of Yazd Traditional Houses, Tehran, 1985. [7] Giyasi J.A., Tabriz (Planning – Architectual Development), Gobustan Journal, No.1, 1988, pp. 42-51. [8] Huseynov F.M., Gahramanova Sh.S. Reconstruction of the city / Tutorial, Baku, Chashioglu, 2004. [9] Huseynov M.A., Bretanitski L.S., Salamzada A.V. History of Azerbaijan Architecture, Moscow, 1963. [10] Memarian, G.H., Introduction to House Typology in Iran, Tehran, University of Science and technology, 1994. ISBN: 978-960-6766-87-9 121 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 Landscape Change in a Mountainous Area in Northeastern Portugal: Implications for Management CÉSAR MOREIRA, JOÃO P. CASTRO, JOÃO AZEVEDO Centro de Investigação de Montanha - CIMO Escola Superior Agrária, Instituto Politécnico de Bragança Campus de Santa Apolónia - Apartado 172, 5301-854 Bragança, PORTUGAL cesar.moreira2005@sapo.pt, jpmc@ipb.pt, jazevedo@ipb.pt Abstract: - In this paper we analyzed changes in landscape composition and structure over the 1958-2005 period in the França Parish (Bragança, northeastern Portugal). Agriculture decreased from 22% to 5% of the area of study. Forests and shrublands became the dominant land uses occupying together 73% of the area of study in 2005. Structurally the landscape became more homogeneous. We speculate that the types of change observed and the tendency for their magnification in the near future will have functional effects on this landscape, namely the creation of condition for the occurrence of catastrophic disturbance events such as wildfires. Key-Words: - Landscape change, landscape management, wildfires, Parque Natural de Montesinho, Montesinho/ Nogueira Natura 2000 Site, Portugal 1 Introduction Landscape change is a fundamental component of landscapes dynamics [1]. Marginal areas in Mediterranean regions have been under landscape change, both in terms of structure and function, as a consequence of land abandonment for a considerable period of time [2]. Although a familiar process in many countries in the Mediterranean basin, in Portugal this is a relatively new phenomenon that became more evident by the end of the XX century. Currently, most of the landscapes in Portugal which are marginal to larger urban centers, namely mountain landscapes, show clear evidence of advanced land abandonment, such as few agriculture areas located around villages, dominance of forest and shrublands land uses and a general homogenization of the landscape. Changes in mountainous areas are critical because they provide ecosystem and landscape services essential to the society as a result of regulation, habitat, production and information functions, i.e. carbon sequestration, water quality and supply, biodiversity and natural resources, just to mention a few. It is therefore important to report and analyze changes in these systems as well as to evaluate their implications on landscape functions, namely on the processes related to the major landscape and ecosystem services, which hasn’t been done in Portugal. It is important in addition, to consider management practices at the landscape level that can contribute to maintain landscape function and dynamics. In this research we analyzed changes over a period of 50 years in landscape composition and configuration of major land use /land cover classes in the França Parish (Bragança County, northeastern Portugal), a typical marginal mountain landscape. Additionally, we infer on possible effects of the observed changes on landscape processes that affect goods and services of these landscapes. 2 Methods 2.1 Area of Study The area of study is the França Parish (Bragança County) located in the North-East of Portugal. This is a 5600-ha area mostly comprised of a granite plateau above 1000m (maximum elevation 1481m). It includes also part of a larger flat valley of metamorphic geology. Average annual precipitation is above 1200mm reaching 1600mm at the higher elevations. Mean annual temperature is 8ºC or less. The Parich includes 3 villages whose population totalled 275 inhabitants (4.9/km2) in 2001 of which only 101 are active. Agriculture (including animal husbandry and forestry) is the most important activity in the parish (40% of the active population) followed by tourism (18%). Within the period of study the populations has decreased from 834 inhabitatants in 1960 to the 275 in 2001. The França Parish is part of the Montesinho/Nogueira Natura 2000 Site (PTCON0002), and part of the Montesinho Natural Park, a protected area established in 1979 by the Portuguese government. ISBN: 978-960-6766-87-9 122 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 2.2 Data We used digitized aerial photographs taken in 1958, 1968, 1980, 1993 (Instituto Geográfico do Exército) and 2005 (Câmara Municipal de Bragança). These coverages were originally obtained at the scales 1:26 000, 1:28 000, 1:30 000, and 1:22 000, respectively. The images were orthocorrected based on ground points. On average we considered 40 points for each photo. The photographs were interpreted based on the COS’2005 land use/land cover system of the Instituto Geográfico Português [3]. Each land unit above 0.25 ha was classified according to the most detailed level of the COS’2005 hierarchical system in a GIS. Field confirmations were done when necessary for the most recent coverages. Landscape structure for each of the dates above was quantified based upon landscape metrics calculated for major land cover classes using FRAGSTATS software [4]. the landscape level. Diversity as measured by the Shannon’s, Simpson’s and Modified Simpson’s indices decreased during that time interval (Table 2) in spite of the new land cover established in the 1990’s and part of the 1993 and 2005 data only. There were also decreases in the Number of Patches and in the Largest Patch Index which represents the percentage of the landscape occupied by the largest patch of all the land classes (Table 3). Edge Density decreased maintaining, however, the initial value of around 166m/ha although there were strong variations within the period of study. Table 2. Measures of landscape diversity for the França Parish from 1958 to 2005. Year 1958 1968 1978 1993 2005 SHDI 1,3509 1,3477 1,3256 1,3662 1,3364 SIDI 0,688 0,6864 0,6607 0,666 0,6551 MSIDI 1,1648 1,1597 1,081 1,0966 1,0645 SHEI 0,754 0,7522 0,7398 0,7021 0,6868 SIEI 0,8256 0,8237 0,7929 0,777 0,7643 MSIEI 0,6501 0,6472 0,6033 0,5635 0,547 3 Results and Discussion 3.1 Composition There were noticeable changes over the 1958-2004 period of study in terms of the major land use classes (Fig.1; Fig. 2; Table 1). Agriculture dropped from 1174ha (22% of the parish area) to 260ha (5%) (Fig.2; Table 1). In the same period forests increased from 741 ha (14% of the area) to 1118 ha (21%). Shrublands is the dominant land cover in any of the dates (Table 1). It represented 47% of the parish area in 1958 and 52.5% of the Parish in 2005. Seminatural pasturelands also increased considerably representing, however, just a residual percentage of the area (Fig. 2; Table 1). Most of the changes described above resulted from abandonment of agriculture areas that became dominated by shrubs in a short period of time. New forest plantations were established mostly in shrublands. Table 1. Major land uses / land covers (percentage of the area) in the França Parish from 1958 to 2005. SHDI-Shannon’s Diversity Índex; MSIDI-Modified Simpson’s Diversity Index; SHEI- Shannon’s Evenness Index; SIEI- Simpson’s Evenness Index; MSIEI-Modified Simpson’s Evenness Index. Table 3. Measures of overall landscape heterogeneity for the França Parish from 1958 to 2005. Year 1958 1968 1978 1993 2005 Number of Largest Patch Patches Index (#) (%) 836 23,9 906 14,6 764 19,1 766 14,9 751 15,7 Edge Density (m/ha) 166,1 170,3 168,3 164,5 166,6 Considering the major land uses/land covers individually (Figure 3) agriculture decreased slightly in Number of Patches and strongly in Mean Patch Area, Largest Patch Index and Edge Density (Figure 3). Forests and shrublands decreased in Number of Patches and increased strongly in Mean Patch Area. The Largest Patch Index increased for forests and decreased for shrublands. Edge Density for these classes increased slightly during the 1958-2005 period. In summary, forests and shrublands became more aggregated in the landscape therefore creating fewer but larger homogeneous landscape units. Agriculture became extremely fragmented and located in the vicinity of villages or in very high fertility sites. 3.2 Landscape Structure Landscape structure in the França Parish also suffered relevant modifications over the nearly 50-year period. Landscape metrics considering major land use classes indicated that there was a decrease in heterogeneity at 3.3 Effects of Landscape Change Considering the trend in landscape change observed for the França Parish with the trend in population shown in Table 4, it is reasonable to admit that forests and shrublands will increase further their dominance ISBN: 978-960-6766-87-9 123 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 in the landscape and that landscape structure will also reflect this dominance by these two major classes. For the most recent date of the time series, these land classes together represented over 73% of the landscape (Table 5; Figure 4). The landscape pattern of these two classes together was becoming more aggregated and the tendency is for aggregation to increase (Table 5; Figure 4). Table 4. Population in the França Parish for the 19602001 period. Year Total Population Active Population 1960 1970 1981 1991 2001 834 714 609 331 275 283 248 173 150 101 high risk for the occurrence of intense and large scale wildfires. Additionally, vegetation encroachment, particularly in younger forest stands, will further increase the fuel in the landscape that with the high connectivity of flammable land uses will favor large scale intense wildfires. Several key functions can be affected by these disturbances, namely hydrological (flooding, soil loss, water contamination), biological (habitat impoverishment; landscape simplification), and cultural (aesthetics). Table 6. Wildfires in the França Parish from 1990 to 2006. Number 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 4 3 0 0 6 6 18 1 25 14 17 28 10 12 3 13 1 Área (ha) 969,5 46,8 0 0 1631,7 213,3 553,4 57,5 1617,1 63,0 1282,7 237,7 55,6 227,0 38,1 58,6 15,0 (%) 17,3 0,8 29,1 3,8 9,9 1,0 28,9 1,1 22,9 4,2 1,0 4,1 0,7 1,0 0,3 Table 5. Structure of shrublands and forest considered together from 1958 to 2005 as indicated by selected landscape metrics Year 1958 1968 1978 1993 2005 Percentage of Landscape (%) 61,2 62,9 67,1 70,5 73,2 Number of Largest Patch Edge Patches Index Density (#) (%) (m/ha) 1126 53,0 136,6 1161 58,0 137,7 930 64,1 130,8 870 67,6 123,3 841 70,3 117,1 This pattern is theoretically favorable to some landscape functions such as hydrological functions, particularly in terms of water quality, and soil loss given the role of shrublands and forest stands in decreasing runoff. In terms of water yield there is a potential decrease at the watershed level since large biomass vegetation types (forests and shrublands) will increase evapotranspiration and decrease water yield. In terms of biodiversity, change of the type described will be favorable, at lease for a certain period of time, to large mammals both in terms of cover and landscape connectivity it provides. However, since most of the species in Mediterranean landscapes are dependent to a certain level of open and/or cultivated areas [2], the changes observed in this work are likely to cause a decrease in species richness and diversity in the area, stronger after abandonment of the few agriculture fields still spread over the landscape. A major reason of concern is the potential this trend represents for the occurrence of catastrophic disturbance events such as wildfires. Landscape change of the same type (but less pronounced) observed in another region of Portugal were predicted to resulted in a higher fire risk in the landscape [5]. The França Parish already presents a high or very high fire frequency (Table 6) and the occurrence of relatively large wildfires. For favorable temperature, moisture, and wind the landscape will show a very One of the advantages of conducting landscape change studies is the possibility of avoiding some of their negative effects. Landscape management is therefore a requirement in marginal lands such as the França Parish in order to prevent the tendency of landscape homogenization and to prevent the development of a landscape structure that is less favorable to large scale fires. 4 Conclusion During the 1958-2005 period, the França Parish went through important changes. There was a strong decrease in agriculture and a regular increase in forests and shrublands. The resulting landscape is less diverse and more aggregated. Considering the population tendency it is likely that the landscape change tendency will be maintained. We expect that these changes are likely to cause change in terms of landscape functioning and in terms of hydrological and biological processes. The major functional change, however, is the possible increase in proneness to fire which can have significant impacts on the ecosystem and landscape services and goods. References: [1] Forman, R.T.T. & M. Godron. Landscape Ecology. John Wiley & Sons, 1986. ISBN: 978-960-6766-87-9 124 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 [2] Farina, A. Principles and Methods in Landscape Ecology. Towards a Science of Landscape. Springer, 2006. [3] Caetano, M., V. Nunes & A. Araújo. Manual da Carta de Ocupação do Solo de 2005 Para Portugal Continental. Instituto Geográfico Português. 2006. [4] McGarigal, K., & B.J. Marks. FRAGSTATS: spatial pattern analysis program for quantifying 1958 1978 landscape structure. Gen. Tech. Report PNWGTR-351, USDA Forest Service, Pacific Northwest Research Station, Portland, OR, 1995. [5] Moreira A, F.C. Rego & P.G. Ferreira. Temporal (1958–1995) pattern of change in a cultural landscape of northwestern Portugal: implications for fire occurrence. Landscape Ecology, 16, 2001, pp. 557–567. 2005 Figure 1. Land use / land cover in the França Parish in three dates of the period of study. 3000 2500 2000 area (ha) 1500 1000 500 0 1958 1968 1978 year 1993 2005 Shrublands Forests Rock Outcrops Semi-natural Pasturelands Agriculture Urban Water Bodies Figure 2. Changes in area for major land uses / land covers between 1958 and 2005 in the França Parish. ISBN: 978-960-6766-87-9 125 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 450 400 12 10 350 Number of patches 300 250 200 150 100 Mean Patch Area (ha) 1958 1968 1978 year 1993 2005 8 6 4 2 50 0 0 1958 1968 1978 year 1993 2005 30 160 140 25 120 Largest Patch Index (%) 20 Edge Density (m/ha) 1958 1968 1978 year 1993 2005 100 80 60 40 15 10 5 20 0 0 1958 1968 1978 year 1993 2005 Forests Shrublands Agriculture Pasturelands Figure 3. Landscape structure change over the 1958 - 2005 period according to landscape metrics for the major general land uses Other Forests and shrublands Figure 4. Change in shrublands and forest considered together over the 1958-2005 period of study. ISBN: 978-960-6766-87-9 126 ISSN 1790-2769 WSEAS International Conference on URBAN PLANNING and TRANSPORTATION (UPT'07), Heraklion, Crete Island, Greece, July 22-24, 2008 Author Index Abdolhoseyni, J. Ahmad, F. Alesheikh, A.A. Ali, Z.M. Azevedo, J. Behzadi, S. Camil, C. Castro, J.P. Catalin, C. Chacha, S. D’ambrosio, S. Dastgerdi, W.Y. De Jong, T.M. Guadagnuolo, D. Guarnaccia, C. Guida, M. Ismail, W.R. Jha, M.K Jones, D. Karlaftis, M. Karri, G. 116 87 23, 28 87 122 28 56 122 56 105 96 83, 101 11, 17 96 96 96 87 105, 110 110 74 105 Khodaparasti, S. Lin, Y.-C. Maleki, H.R. Maricic, T. Milioti, C. Mohammad, R. Moreira, C. Mustafa, Z. Ou-Yang, C. Peng, N.Y. Poorazizi, E. Quartieri, J. Spyropoulou, I. Suradi, N.R. Tejada, J.D. Tsouchlaraki, A. Udenta, F. Wey, W.-M. Wu, K.-Y. Zekovic, S. Zoaki, E. 69 62 69 33 74 87 122 87 62 87 23 96 74 87 39 47 105 62 62 33 47 ISBN: 978-960-6766-87-9 127 ISSN 1790-2769