Please note that the papers from this proceedings are not available online, only from the CD itself.

 book references
 full text searching
keyword index 
author index 

Conference Programme

Friday, July 2 Saturday, July 3 Sunday, July 4 Monday, July 5
7:00-9:00 breakfast 7:00-9:00 breakfast 7:00-9:00 breakfast
9:00-10:30 session 1
1: Opening Session
9:00-10:30 session 5
5: Viualisation for Design and Decision Support 2
9:00-10:30 session 9
9: Agent Technology 2
10:30-11:00 coffee break 10:30-11:00 coffee break 10:30-11:00 coffee break
11:00-12:30 session 2
2: Applications of Artificial Intelligence
11:00-12:30 session 6
6: Agent Technology 1
11:00-12:30 session 10
10: Collaborative Design
12:30-14:00 lunch break 12:30-14:00 lunch break
14:00-15:30 session 3
3: Viualisation for Design and Decision Support 1
14:00-15:30 session 7a+b
7a: Geographical Information Systems
7b: Knowledge Representation
12:30 closure
15:30-16:00 coffee break 15:30-16:00 coffee break
16:00-18:00 registration
18:30 reception
Reception and Buffet
16:00-18:00 session 4a+b
4a: Spatial Simulation
4b: Design Research and Support
16:00-18:00 session 8a+b
8a: Models for Urban Planning
8b: Building Services and Performance
19:00 dinner 19:00 dinner


Session Details

This page contains the abstracts of all papers included in the conference proceedings. You can use your browser's search facility to search for specific words in the abstracts.

RA / D
1 Opening Session
Harry Timmermans
Eindhoven University of Technology
Introduction to the Conference
Jos van Leeuwen
Eindhoven University of Technology
Stefania Bandini, Sara Manzoni, and Giuseppe Vizzari
Department of Computer Science, Systems and Communications, University of Milan-Bicocca, Italy
keywords: Multi-Agent Systems, 3D modelling, Simulation.
abstract: The paper introduces a Multi Agent Systems (MAS) approach to crowd modelling and simulation, based on the Situated Cellular Agents (SCA) model. This is a special class of Multilayered Multi Agent Situated System (MMASS), exploiting basic elements of Cellular Automata. In particular SCA model provides an explicit spatial representation and the definition of adjacency geometries, but also a concept of autonomous agent, provided with an internal architecture, an individual state and behaviour. The latter provides different means of space-mediated interaction among agents: synchronous, between adjacent agents, and asynchronous among at-a-distance entities. Heterogeneous entities may be modelled through the specification of different agent types, defining different behaviours and perceptive capabilities. After a brief description of the model, its application to simple crowd behaviours will be given, and an application providing the integration of a bidimensional simulator based on this model and a 3D modelling application (3D Studio) will also be described. The adoption of this kind of system allows the specification and simulation of an architectural design with reference to the behaviour of entities that will act in it. The system is also able to easily produce a realistic visualization of the simulation, in order to facilitate the evaluation of the design and the communication with involved decision-makers. In fact, while experts often require only abstract and analytical results deriving from a quantitative analysis of simulation results, other people involved in the decision-making process related to the design may be helped by qualitative aspects better represented by other forms of graphical visualization.
Tomo Cerovsek and Bob Martens
University of Ljubljana; Vienna University of Technology
keywords: Scientific Knowledge Management, Retrospective CAAD Research, Graph Theory, CAAD-Related Publications, Web-Based Bibliographic Database
abstract: This paper discusses the extended use of the Cumulative Index of CAAD (CUMINCAD) - a digital library set up in 1998 serving the CAAD-community as an important source of scientific information with over 6.000 recorded entries published on-line. The aim of this paper is to elaborate a related Citation Index to CUMINCAD - with over 20.000 references - and to provide information on entries with an exceptional high impact in the CUMINCAD database. The importance is determined through its use (citing) in the framework of afterwards published scientific materials. By utilizing graph theory methods extensive citation analyses will be presented illustrating the impact of particular contributions in different research topics.
2 Applications of Artificial Intelligence
Ertan Yesilnacar and Gary J. Hunter
The University of Melbourne, Australia
keywords: Landslide Susceptibility Mapping, Neural Networks, Spatial Decision Support Systems.
abstract: Landslides are a major natural hazard in many areas of the world, and globally they cause hundreds of billions of dollars of damage, and hundreds of thousands of deaths and injuries each year. Landslides are the second most common natural hazard in Turkey, and the Black Sea region of that country is particularly affected. Therefore, landslide susceptibility mapping is one of the important issues for urban and rural planning in Turkey. The reliability of these maps depends mostly on the amount and quality of available data used, as well as the selection of a robust methodology. Although statistical methods generally have been implemented and used for evaluating landslide susceptibility and risk in medium scale studies, they are distribution-based and cannot handle multi-source data that are commonly collected from nature. These drawbacks are responsible for the on-going investigations into slope instability. To overcome these weaknesses, the desired technique must be able to handle multi-type data and its superiority should increase as the dimensionality and/or non-linearity of the problem increases – which is when traditional regression often fails to produce accurate approximations. Although neural networks have some problems with the creation of architectures, processing time, and the negative “black box” syndrome, they still have an advantage over traditional methods in that they can deal with the problem comprehensively and are insensitive to uncertain data and measurement errors. Therefore, it is expected that the application of neural networks will bring new perspectives to the assessment of landslide susceptibility in Turkey. In this paper, the application of neural networks for landslide susceptibility mapping will be examined and their performance as a component of spatial decision support systems will be discussed.
Alexandra Akamine and Antônio Nélson Rodrigues da Silva
University of São Paulo, School of Engineering of São Carlos, Brazil
keywords: Artificial Neural Networks, Spatial Interaction Models, Education Infrastructure.
abstract: One of the serious problems faced by the Brazilian municipalities is the scarcity of resources for building education infrastructure. This asks for an optimal allocation of the available resources that includes, among other things, a rational spatial arrangement of the supply points (i.e., schools) in order to increase the demand coverage (i.e., students). If it is possible to foresee the regions where the demand is going to be concentrated, it is then possible to plan the location of new facilities and to assess the impact on the future level of service of the entire system. Considering that one of the consequences of the location-allocation process is the distribution of trips from demand points to supply points throughout the city, therefore affecting the overall intraurban accessibility conditions to essential services such as education, there is a strong need of models that planners can rely on to predict the future trip distribution patterns. As a result, the objective of this work was to evaluate the performance of Artificial Neural Networks (ANN) when applied to spatial interaction models, the so-called Neural Spatial Interaction Models. This was done in a practical context, in contrast to the more theoretical works commonly found in literature. The practical application showed that the neural spatial interaction model had different performances when compared to the traditional gravity models. In one case the neural models outperformed the gravity models, while on the other case it was just the opposite. The explanation for this may be in the data or in the ANN model formulation, as discussed in the conclusions.
Lidia Diappi, Paola Bolchim, and Massimo Buscema
Department of Architecture and Planning – Polytechnic of Milan, Italy; Semeion Centre – Rome, Italy
keywords: Neural Networks, Self-Organizing Maps, Land-Use Dynamics, Supervised Networks.
abstract: It is widely accepted that the spatial pattern of settlements is a crucial factor affecting quality of life and environmental sustainability, but few recent studies have attempted to examine the phenomenon of sprawl by modelling the process rather than adopting a descriptive approach. The issue was partly addressed by models of land use and transportation which were mainly developed in the UK and US in the 1970s and 1980s, but the major advances were made in the area of modelling transportation, while very little was achieved in the area of spatial and temporal land use. Models of land use and transportation are well-established tools, based on explicit, exogenously-formulated rules within a theoretical framework. The new approaches of artificial intelligence, and in particular, systems involving parallel processing, (Neural Networks, Cellular Automata and Multi-Agent Systems) defined by the expression “Neurocomputing”, allow problems to be approached in the reverse, bottom-up, direction by discovering rules, relationships and scenarios from a database. In this article we examine the hypothesis that territorial micro-transformations occur according to a local logic, i.e. according to use, accessibility, the presence of services and conditions of centrality, periphericity or isolation of each territorial “cell” relative to its surroundings. The prediction capabilities of different architectures of supervised Neural networks are implemented to the south Metropolitan area of Milan at two different temporal thresholds and discussed. Starting from data on land use in 1980 and 1994 and by subdividing the area into square cells on an orthogonal grid, the model produces a spatial and functional map of urbanisation in 2008. An implementation of the SOM (Self Organizing Map) processing to the Data Base allows the typologies of transformation to be identified, i.e. the classes of area which are transformed in the same way and which give rise to territorial morphologies; this is an interesting by-product of the approach.
3 Viualisation for Design and Decision Support 1
C. Pettit, A. Nelson, and W. Cartwright
RMIT University, Melbourne, Australia
keywords: Scenario Planning, Geographical Visualisation, GIS, Multimedia, VRML.
abstract: This paper examines the development of a prototype suite of on-line integrated multimedia-GIS tools to assist in bottom up decision-making. These tools are being developed in the context of scenario planning to enable the community to actively explore different land use options and the implication of government structure and strategic plans. A case study approach is undertaken, focusing on the Jewell Station Neighbourhood, situated in the City of Moreland, Greater Melbourne Region, Australia. The paper documents the first stage of the project, in developing three land use scenarios delivered through a range of technologies including: VRML, HTML, GIS, Pixmaker and Flash. The paper concludes by outlining the future directions of this research that include: the construction of a virtual sandbox, usability testing, and community consultation.
Jack Barton, B. Parolin, and V. Weiley
The University of New South Wales, Australia
keywords: Spatial Decision Support System, Public Housing, Community Renewal, Security, Open Source, Interoperability, Visualisation, 3D GIS, PPGIS, X3D.
abstract: This paper is reporting on a research project undertaken jointly between the University of New South Wales (UNSW) and the NSW Department of Housing (DoH) to develop a Spatial Decision Support System (SDSS) to assist planning, management and evaluation in areas of high public housing concentration. In the paper we will describe the development of the SDSS, the specific spatial problems challenging the DoH and the potential for the system to incorporate a range of social, financial and physical data, both internal and from other sources, for interaction and presentation in a three dimensional environment. The prototype SDSS attempts to address the specific challenges of providing better service for clients of the DoH. An information audit and survey has been conducted of the department’s resources and needs. Issues identified include the management of high-rise and superlot areas, crime mapping, community interactivity, internal and intergovernmental information sharing, interoperability and maintaining confidentiality and security of data. Interactive 3D visualisation of the model is facilitated by use of the 3map free geospace platform. Use of open source code and open standards such as X3D for 3D graphics interchange allow the project to explore advanced visualisation techniques while ensuring interoperability and data longevity.
Z.J. Shen and M. Kawakami
Faculty of Engineering, Kanazawa University, Japan
keywords: WEBGIS, JAVA3d, Planning restrictions, Public participation.
abstract: The district plan in Japan is designed and implemented through public participation, which is based on the zone restrictions of land use system. The usable space of a building can be generated according to the zone restrictions implemented in a district plan, which include the planning controls of high altitude, oblique line and so on. Residents can choose control items as their preference in order to control the urban physical environment of the district. In this paper, we discuss how to generate the virtual world according to the data set of GIS including planning control items and how it can work for net participation. In the future, we would like to conduct an Internet social experiment with the cooperation of a city government and residents in order to support the district plan in a local city.
4a Spatial Simulation
Dimitris Ballas, Richard Kingston, and John Stillwell
Department of Geography, University of Sheffield, England; School of Planning and Landscape, University of Manchester, England; School of Geography, University of Leeds, England
keywords: Spatial Microsimulation, Spatial Decision Support Systems, Geotools.
abstract: This paper discusses the potential of a spatial microsimulation-based decision support system for policy analysis. The system can be used to describe current conditions and issues in neighbourhoods, predict future trends in the composition and health of neighbourhoods and conduct modelling and predictive analysis to measure the likely impact of policy interventions at the local level. A large dynamic spatial micro-simulation model is being constructed for the population of Leeds (approximately 715,000 individuals) based on spatial microsimulation techniques in conjunction with a range of data, including 2001 Census data for Output Areas and sample data from the British Household Panel Survey. The project has three main aims as follows: (i) to develop a static microsimulation model to describe current conditions in Leeds; (ii) to enable the performance of ‘What if?’ analysis on a range of policy scenarios; and (iii) to develop a dynamic microsimulation model to predict future conditions in Leeds under different policy scenarios. The paper reports progress in meeting the above aims and outlines the associated difficulties and data issues. One of the significant advantages of the spatial microsimulation approach adopted by this project is that it enables the user to query any combination of variables that is deemed desirable for policy analysis. The paper will illustrate the software tool being developed in the context of this project that is capable of carrying out queries of this type and of mapping their results. The decision support tool is being developed to support policy-makers concerned with urban regeneration and neighbourhood renewal.
A. Ohgai, Y. Gohnai, S. Ikaruga, M. Murakami, and K. Watanabe
Toyohashi University of Technology, Toyohashi, Japan; Yamaguchi University, Yamaguchi, Japan; Kogakuin University, Tokyo, Japan; The University of Tokushima, Tokushima, Japan
keywords: Community-Based Planning for Disaster Mitigation, Fire Spreading, Fire Fighting Activity, Cellular Automata Modeling, Planning Support Tool.
abstract: As a tool to support collaboration in community-based planning for disaster mitigation in Japanese old wooden built-up areas, we attempt to develop a fire spreading simulation model incorporated a fire fighting activity using Cellular Automata (CA). The proposed model can deal with the process of fire spreading in a building that traditional models can not represent. Whether or not fire can spread is based on a stochastic calculation process to reproduce uncertain fire spreading. The errors caused by the stochastic factor are analyzed by carrying out simulation two or more times under the same condition. Moreover, the reproductivity of the model is examined by comparing simulation results with actual fire records.
Kay Kitazawa and Michael Batty
Centre for Advanced Spatial Analysis, University College London
keywords: Pedestrian Modelling, Retail Movement, Shortest-Path, Genetic Algorithms
abstract: This paper introduces a study of pedestrian behaviour modelling which incorporates ideas about agent-based systems and the traffic models based on the utility-maximization theory. The aim of this study is to examine the applicable scope of one of the basic assumptions in existing spatial behaviour models; the route with shortest distance maximizes the utility of each pedestrian’s travel. Although shortest-path models have been widely used in the field of Traffic management to predict routing behaviour, there can be seen a lot of erratic behaviour in urban areas, shopping migration behaviour for instance, which can not be explained by them. Thus, it is important to identify other possible influential factors on their utility maximization process in order to develop more explicable models of pedestrian movements. In this study, we implemented a simulation model using the shortest-path model as one of evaluation criteria of Genetic Algorithms (GA) to computationally emulate retail movements of shoppers in a big shopping centre and to test the accuracy of the model by comparison between the routes estimated by the model and actual trajectories of shoppers. This simulation system will be used as a platform for further modelling.
Kuo-Chung Wen and Wei-Lung Chen
Graduate Institute of Architecture and Urban Planning, Chinese Culture University

keywords: Genetic Algorithms, Decision Support System, GIS, Evacuation Path
abstract: Because of the special geography features and subtropics weather in Taiwan, we need to provide correct information to help people making decision when they are in disaster. So the disaster decision support system must offer proper information of evacuation path to people. This research has shown the difficulties associated with the GIS and the flooding evacuation path search through the huge searching space generated during the network analysis process. This research also presents an approach to these problems by utilizing a search process whose concept is derived from natural genetics. Genetic algorithms (GAs) have been introduced in the optimization problem solving area by Holland (1975) and Goldberg (1989) and have shown their usefulness through numerous applications. We apply GA and GIS to choice flooding evacuation path in metropolitan area in this study. We take the region of Shiji city in Taiwan for case. Firstly, we establish the node relationship of GA calculation, the level of the weight is the standard of the date that is exported by Disaster Database. Secondly, we apply GA to calculate different evacuation path in different time series. Finally, we build the model of choosing flooding evacuation path.
4b Design Research and Support
S. Kacher, J.-C. Bignon, and G. Halin
MAP-CRAI – Architecture and landscape Modeling - Research Center in Architecture and Engineering, Architecture School of Nancy, France
keywords: Weight Given to Terms, Image Describers, Semantic Description, Visual Criteria.
abstract: Architectural design is a domain where using pictures (e.g., drawing, photographs, …) is essential because the nature of the information transmitted by photographic image is often easier to interpret. The fact is that an image requires less interpretation than a text. The information transmitted by image (element shape, colour, light, ...) is already “put in shape” and so can be more easily integrated into the design process. This paper presents a way to index more efficiently an image database of the wooden architecture domain. Images in our databases illustrate real architectural elements. This work aims to analyse the representation of the real element illustrated by images. The analysis will allow us to identify some criteria related to the visual features of each image. The identified criteria will be used in a discriminating way to associate a weight with an indexation term describing its representation illustrated by an image. The importance of that representation (according to what is seen at first) is evaluated depending on graphic rules which correspond to the graphic properties of the representation of the element in each image.
M.F.Th. Bax and H.M.G.J. Trum
Eindhoven University of Technology, The Netherlands
keywords: Levels, Hierarchy, Architecture, Composition, Complexity, Control.
abstract: The notion of Level (in a scale of Levels) is probably the most authentic notion in Architecture. Already in the work of Vitruvius the notion is implicitly present in the triad ‘ordinatio – symmetria – eurythmia’. In more recent times, the notion always appears in relation with hierarchical organization as a means of control of quality. However used in drawings and in architectural discourse, the term lacks precision; there are many types of level like abstraction, specification, dependency, resolution levels etc., but no operational definition can be found as a notion that structures architectural objects and design processes simultaneously in a consistent way. Defining this notion of Level is the purpose of this paper. An example of application in an architectural decision-making process completes the paper.
Jonas Lindekens and Ann Heylighen
Vrije Universiteit Brussel, Belgium; Katholieke Universiteit Leuven, Belgium
keywords: Design Process, Design Method, Design Strategy, Re-Design, Design Tool
abstract: Addressing the issue of building re-use, this paper reports on the development of a re-design tool, which aims at helping students and novices (and even experienced architects) in understanding design decisions that have steered existing re-design projects. Furthermore the tool tries to support designers in using this understanding in their own design projects. The paper starts by situating the re-design tool within the overall research setup, and continues by motivating the observations that have led to developing this tool. The core of the paper describes the re-design tool itself and illustrates it with a concrete example. Finally, the tool is framed within the wider context of (architectural) design research, so as to explore where the tool accords or collides with generally accepted design theories.
Jonas Lindekens
Vrije Universiteit Brussel, Belgium
keywords: Architectural Re-Design, Design Process, Case-Based Design, Reduction Mechanisms.
abstract: Observation of the design process of an architect shows that, while building up an argumentation for taking a design decision, different mechanisms of data transformation are used. The paper argues that this transformation is a key element in understanding architectural design processes. A theoretical description of these mechanisms forms the framework to discus a sequence of design decisions derived from a real-world design situation. After outlining how this can be implemented in a case-based design supporting tool, the paper concludes with a discussion of advantages and downsides the use of the tool might entail.
5 Viualisation for Design and Decision Support 2
Ji-Hyun Lee and Wei Qian
Graduate School of Computational Design, National Yunlin University of Science & Technology, Taiwan (R.O.C.)
keywords: Design Support System, Sensibility Ergonomics, Color Coordination, Color Image.
abstract: Color selection plays a vitally important role in creating impressions of individuals or companies because colors have sensibility aspects and relate to some images or associations. Based on both the theory of color harmony and the sensibility ergonomics, some quantitative and systematic researches on the color image have been developed. In this paper, we suggest a color coordinate system that supports the color analysis and the color harmony functions using color images, which can be captured by corresponding adjective words. We focus on a system prototype for interior design domain to exemplify our concepts in this paper, even though this system can be applied for all design domains.
Remo Burkhard
Swiss Federal Institute of Technology
keywords: Decision Making, Knowledge Transfer, Visualization Types, Interfunctional Communication, Business Knowledge Visualization, Information Visualization
abstract: The transfer of knowledge between planners and business decision makers can be improved when planners combine traditional visualizations with business knowledge visualizations. Today architects and urban planners use visualization methods such as sketches, diagrams, drawings, renderings, models and animations to illustrate their projects. While spending an enormous amount of time to illustrate a project, almost no time is used to illustrate business relevant information that decision makers need (i.e., revenue models, risks, return on investments, project phases). Consequences are information overload, misinterpretation or even misuse of information. Juxtaposing the visualizations that planners and decision makers use reveals a major gap: Both groups use different visualization types and are not familiar with the visualization types of each other. This paper stresses the importance to expand the visualization types of planners with business knowledge visualizations. First, it discusses the functioning of visual representations for the transfer of knowledge. Second, it introduces a general knowledge visualization framework. Third, it illustrates examples from an innovative office that improved knowledge transfer with decision makers in urban planning projects. We found that combining traditional visualizations with business knowledge visualizations reduces the information overload, prevents misinterpretation, increases the information quality, improves communication and as a consequence improves decision making. We found that decision makers pay extra for these visualization types, which therefore is a new source of income for planners. The results have implications for the education of future architects.
Roland Göttig, Joanna Newton, and Stefan Kaufmann
Technische Universität München, Germany
keywords: 3D-Systems, Virtual Reality, Powerwall, Holobench, HMD, CAVE, User Interfaces, Visual Display Qualities, Design Process.
abstract: Contemporary advanced virtual reality systems use different stereoscopic 3D visualization technologies. In this study, VR systems from one projection wall up to VR systems with six projection walls have been evaluated. Besides the optical properties tested with architectural 3D models, the user interfaces have been analyzed with reference to exact and intuitive control abilities. Additionally, the workflow of an early architectural design process with CAAD generated 3D models and VR visualization techniques was analyzed. It turns out that current VR systems exhibit shortcomings in visual and spatial representations, as well as tools for an early design process.
6 Agent Technology 1
V. Tabak, B. de Vries, and J. Dijkstra
Eindhoven University of Technology, The Netherlands
keywords: Building Simulation, Decision Support Systems, User Behaviour, Petri-nets, Activity Based Modelling
abstract: The aim of the proposed project is to develop methods for the simulation of space utilisation. Up to now no methods for building performance evaluation are available which involve the occupants of the building. Instead, assumptions are made about people’s movement through space and their responses to the environment. These assumptions are input for important design decisions (e.g. capacity of elevators, width of corridors, escape routing) sophisticated calculations (e.g. cooling and lighting calculations) and simulations (e.g. airflow simulation, evacuation simulation). Reliable data on human movement are very scarce and can be valuable input to research in other research areas. New computer technologies allow for dynamic simulations that will provide insight into the building to be built. The research project builds upon existing methods that need to be tailored and/or extended to apply them to the building domain and to support real-time simulation.
Dick Saarloos, Theo Arentze, Aloys Borgers, and Harry Timmermans
Eindhoven University of Technology, The Netherlands
keywords: Multi-Agents, Local Land-Use Planning, Alternative Plan Generation, Reasoning, and Interaction
abstract: This paper addresses the MASQUE multi-agent framework for generating alternative plans in local land-use planning, In this framework agents represent land-use experts and initiate the development of plan proposals and request each other to express their claims in order to incrementally draw up these proposals. Presented is a probabilistic approach to the implementation of those agents to enable them to make decisions under uncertainty. It is described what personal and collective beliefs they construct and use in order to strategically choose their actions. Negotiation takes place between the initiating agent and the others in order to reach agreement on the incorporation of the claims. The negotiation is organized as an iterative process in which both parties consider conciliatory adjustments of their strategies, and thus their decisions, in order to try to find mutually accepted solutions.
O.T.J. Devisch, H.J.P. Timmermans, T.A. Arentze, and A.W.J. Borgers
Eindhoven University of Technology, The Netherlands
keywords: Micro Simulation, Multi-Agent Systems, Spatial Simulation Models, Scenario Analysis, Heuristic Decision-Making, Location-Choice Behaviour, Decision-Analysis Trees, Residential Mobility, Lifetime Utility.
abstract: SwarmCity is being developed as a micro-simulation model, simulating the location-choice behaviour of a population of households, retailers, firms, developers, etc. reacting to an urban plan. The focus of SwarmCity lies –in a first phase- on the decision-making procedures of households, conceptualised as a series of three processes: awakening, search and choice. The methodology used to implement these processes makes use of life-time utility and decision-analysis trees. The final model should work as a scenario-analysis tool, allowing planners, developers, retailers and municipalities to test intervention-proposals, to evaluate legislations, to measure the attractiveness of services, to quantify residential mobility, etc.
7a Geographical Information Systems
Daniel S. Rodrigues, Léa C.L. Souza, and José F.G. Mendes
University of Minho – Department of Civil Engineering, Braga, Portugal; State University of São Paulo – Department of Architecture, Urbanism and Landscape, Bauru, São Paulo, Brazil
keywords: Sky View Factor, Urban Geometry, GIS Extension, Urban Heat Island.
abstract: This paper presents a second version of the 3DSkyView extension. The purpose of that extension was to implement a calculation algorithm for assessment and visualization of sky view factors (SVF) by means of tools available in a Geographical Information System (GIS). The sky view factor is a thermal and geometric parameter pointed out in the specialized literature as one of the main causes of urban heat islands. A 3D-GIS is a powerful tool for reaching the goal of this research because it allows the storage, treatment and analysis of tri-dimensional urban data, in addition to a high level of flexibility for incorporating calculation algorithms. The objective in the 3DSkyView extension is to optimize the determination of that factor, not only reducing its demanding calculation and graphical representation time, but also generating a simplified tool for replacing expensive photographic equipment usually applied on this matter. Enhancing functions of ArcView GIS 3.2, the first version of that extension showed a very good performance allowing the automatic delineation and determination of SVF. That performance was although limited to a single observer point. The simulation of SVF for several view points in urban canyons was only possible by applying the extension as many times as the number of observers considered. Therefore, this second version was now developed in order to allow simultaneous determination of SVF for many view points. In addition, the 3DSkyView new interface is more flexible, in a way that the user may choose the kind of output wanted (graphical and/or tabular). With this new feature it is then easier to create a continuous SVF map for an entire area.
Masahiro Murakami, Kotaro Higuchi, and Akihiro Shibayama
Kogakuin University, Japan
keywords: Convenience Store Robbery, GIS, Graph Theory, Indexes of Graphs and Networks.
abstract: This study focuses on road environment around robbed convenience stores and elucidates the road characteristics that are conducive to convenience store robberies by using GIS and indexes of graphs and networks. The method of this study is as follows: Five convenience store robberies, which occurred in Metropolitan Tokyo, were selected from newspaper reports. Then, road networks within a 1-kilometer radius of the robbed convenience stores were extracted from digital maps with a scale of one-twenty five hundredth (Geographical Survey Institute and Bureau of City Planning Tokyo Metropolitan Government). After adding the road networks and the attributes such as road width, we investigate the road characteristics using GIS and indexes of graphs and networks. Finally, we demonstrate several factors associated with convenience store robberies based on this compiled information.
Alexandra Tisma
Netherlands Institute for Spatial Research, The Hague, The Netherlands
keywords: Map Making, Computational Tools, Interactive Decision Making.
abstract: By developing framework plans on a large scale such as a country or a region, planners use economic prognoses which show future needs for space for new spatial developments expressed in units of surfaces. Planners and designers make drawings and sketches to show where those new areas will be situated, but they do not really measure the surfaces of spaces they created. This often leads to incorrect images which can be wrongly understood by actors in decision making. To avoid this problem the Netherlands Institute for Spatial Research developed mapmaking software called RasterPlan. The purpose of this tool is to make design decisions quantitatively checkable and transparent. The maps which result from the RasterPlan are precise in geographical positioning and defining the surfaces of designed areas. RasterPlan allows realization of a quantitative program for future spatial needs for various functions such as housing, green and water areas working, and recreation. In addition to quantitative calculations, qualitative criteria for location choice can be also expressed in a form of suitability maps or buffers. This paper presents an experimental implementation of RasterPlan for the design of the future development of the Province North Brabant.
7b Knowledge Representation
Ih-Cheng Lai
Graduate Institute of Architecture, National Chiao Tung University, Hsinchu, Taiwan
keywords: Idea Association, Case-Based Reasoning, Retrieval, Adaptation, Protocol Analysis
abstract: This research makes use of a cognitive study to explore a mechanism for associating ideas in a brainstorming session. Firstly, we propose a linking model integrating three principles of idea association (similarity, contrast and contiguity) with two processes of case-based reasoning (retrieval and adaptation). For identifying the types and mechanisms of linkages within the linking model, a design experiment and its protocol analysis was conducted. Finally, a framework for case-based reasoning to support idea association called Dynamic Idea-Maps (DIM) is proposed, and its components and mechanisms are elucidated.
Teng-Wen Chang
National Chiao Tung University, Taiwan
keywords: Design Puzzles, Design Collage, Puzzle-Making, Andragogy, Game Play
abstract: The design process is a puzzle-solving process. Two groups of researches that share many similarities with Puzzle-solving design process are the process of game-playing and playful learning. The main argument is using the “playing” characteristics to amplify and explore the learning process, furthermore the design process. In addition, puzzles imply playful exploration that utilizes the characteristics of “playing a game” as “solving a puzzle”. Puzzle making and puzzle solving provides an incremental exploration mechanism that is more intuitive for design learning. For understanding and realizing puzzles in design learning, this research is divided into two stages of researches—manual design puzzles and interactive design puzzles. By analysing the outcome from manual design puzzles, this research proposes a framework called (interactive) “design puzzles”. The conceptual and implementation framework of this view of design is elaborated in this paper as well as a particular design puzzle called puzzle collage is described as the realization of design puzzles.
Ahmad Rafi, Tran Hoai Nam, Kim Jin Woo, and Wai Lau
Faculty of Creative Multimedia, Multimedia University, Cyberjaya Malaysia; Innotive Resarch Lab, Innotive Corporation, Multimedia University, Cyberjaya Malaysia
keywords: InnotiveBrowser, Digital Asset Management, Multimedia Archiving, VR Technology
abstract: This paper explains the results of a collaborative research project between Multimedia University (Malaysia) and Innotive Corporation (Korea) to manage, design a multimedia archiving system and visualising knowledge for the students or the users in the Faculty of Creative Multimedia, Multimedia University. This research introduces InnotiveBrowser technology, a high performance multimedia display software that enhances the ability of user to search and discover digital content. The unique result of this method is that the images to be viewed are not limited to available RAM, instead the content utilises the available storage directly from the disk (hard drives). In other words, the larger the size of the hard disk, the greater the number of content information can be stored and displayed. This system is employed with Virtual Reality (VR) techniques particularly imparting viewing technology (pixel-on-demand) and navigation strategy to increase the viewing speed of multimedia information in real-time over the Internet, broadband and even via PDA platforms. This research hopes to set the benchmark for multimedia archiving system that can be applied in other CAD, CAAD or most of the design or production-based teaching and learning environment. The early findings of this research have been patented and this paper will demonstrate the research ideas and explain how we implement and customise the technology and content development in the Faculty of Creative Multimedia, Multimedia University.
8a Models for Urban Planning
Lucia Tangari, Michele Ottomanelli, and Domenico Sassanelli
Polytechnic University of Bari
keywords: Contingent Valuation, Transportation Planning, Urban Decongestion
abstract: This study introduces a methodology for the evaluation, among a set of possible plans and/or policies, of a transport system which could reduce the effects of the network congestion on an urban area. Hence it is proposed a procedure which contains a new element compared with the current methodologies that is the final user’s evaluation of the examined system by turning to the application of the contingent valuation method, a technique like stated preferences. The approach we propose is based on the necessity for a determination of an optimal solution to the urban congestion to be established on the politic acceptability by the final user. Solutions which result virtually feasible for a public administration both on a technique and economic point of view often meet hostility by individuals. Consequently, it would be interesting for policy makers to adopt a process of valuation which could let to understand the user’s sensitivity and hostility towards specific configurations of the system, chosen as solution to decongestion (traffic calming, road-pricing, auto-free zones) and consequently to make less unpleasant the strategy to be carried out to control congestion (Harrington et al., 2001). In this way, a user does not judge the implementation of a set of projects made by a decision-maker as imposition and he is willing to pay in order to fulfil the chosen scenario. Finally we describe an application of the proposed methodology relating the definition of the integrated transport system in the metropolitan area of Bari, chief town of Apulia.
Claudia Pelizaro, T.A. Arentze, and H. J. P. Timmermans
Eindhoven University of Technology, The Netherlands
keywords: Design & Decision Support Systems, Integrated Urban Models, Urban Planning, Urban Green Space
abstract: A spatial decision support system for the planning, design and maintenance of urban green space is presented. The objective of the system under development is to assist local authorities and green space administrators to strategically enhance the supply of urban greening with the right type and variety of green space that maximizes public welfare. The system is being developed starting from a modelling perspective and GIS functionalities are added conform the needed analysis and subroutines within the system. The system has been written in the C++ Borland Builder 5 programming environment. GIS capabilities and dynamic mapping are added using MapObjects 2.0.
Fabrizio Fantini, Ugo Schiavoni, and Barbara Barboni
University of Roma “Tor Vergata”
keywords: Urban Planning, Models, Operations Research, Project Management
abstract: We consider program proposals as separate projects and use an incremental decision process (Incremental Functional Formulation) backed by methods of optimisation through incremental and interactive decision processes which have been developed in operations research. These methods fit the ill-structured problems in planning better than the conventional optimisation methods. We'll conclude showing one of the possible concrete applications of our model to a real-life case, where one of the main results was a suggested expense of € 4 millions leading to the 90% of the result possible with the cost of € 10 millions (but with just a 40% expenditure).
Maurício Polidori and Romulo Krafta
Rio Grande do Sul Univerity – Ecology Centre – Brazil
keywords: Environment, Urban Growth, Urban Morphology, Simulation
abstract: This work presents the synthesis of a model of urban growth dedicated to accomplish simulations of urban spatial dynamics, based on integrated urban and environmental factors and promoting simultaneity among external and internal growth. The city and surrounding environment are captured and modeled in computational ambient, by application of the centrality / potential model (Krafta, 1994 and 1999), with support of graph theory, cellular automata, GIS and geocomputation. The model assumes the city as a field of opportunities for obtaining income, mediated by the space, which is composed of urban and environmental attributes, that work as attractors or as resistances for the urban growth. The space configuration and the distribution of those attributes generate tensions that differentiate qualitatively and quantitatively the space, through the centrality measure (built with the support of graphs techniques), coming to provoke growth in places with larger potential of development (built with the help of techniques of CA – cellular automata). Growths above environmental thresholds are considered problems, generated and overcome in the same process of production of the urban space. Iterations of that process offer a dynamic behaviour to the model, allowing to observe the growth process along the time. The model presents several possibilities: a) urban - natural environment integration; b) internal and external growth integration; c) variety in the scale; d) GIS integration and geocomputation; e) user interface; f) calibration; g) theoretical possibilities; and h) practical possibilities.
8b Building Services and Performance
G. Bi and B. Medjdoub
School of Built Environment, University of Nottingham, UK
keywords: Case-Based Reasoning, Constraint Satisfaction Problem, Ceiling Voids Layout, Complex Geometry, Large Problem.
abstract: In this paper an object-based CAD programming is used to take advantage of standardization to handle the schematic design, sizing, layout for services in a building ceiling void. From the specification of the building 3D model, our software proceeds through different steps; from the determination of the standard number and size of fan coils to the generation of 3D solutions. In order to deal with more complex geometry and larger problems, we have used a hybrid approach: Case Based Reasoning (CBR) within Constraint Satisfaction Problem (CSP) approaches. In practice, engineers in building services use previous solutions and adapt them to new problems. CBR mirrors this practical approach and does help us to deal with increasingly complex geometry effectively, and meanwhile CSP has been used for layout adaptation. The results have shown that it is possible to define and implement standard solutions to produce designs comparable with current practice. The benchmarking exercise has underlined many advantages and made some suggestions for further development. This project is funded by The Engineering and Physical Sciences Research Council (EPSRC) in UK.
Johanna E.M.H. van Bronswijk, Ludovicus G.H. Koren and Cor E.E. Pernot
Eindhoven University of Technology, The Netherlands; TNO Building and Construction Research, Eindhoven, The Netherlands
keywords: Health Prediction, Built Environment, Epidemiology, Modelling
abstract: The influence of built environments on vitality and productivity of users is paramount. Since the introduction of Industrial, Flexible and Demountable Building, domotics, smart buildings, in general: mass-produced, intelligent and learning built environments, tailored built environments are within reach. This has resulted in the need for methodologies to predict short-term and long-term health effects of different built-environment constellations. Epidemiology has developed and validated methods to assess changes in prevalence of inflictions and other unhealthy conditions, as well as the number of healthy and vital years in a life span. After analysing the relationships among building (services) parts and its combinations, health determinants (exposures) and health outcomes, we could adapt the healthy years assessment (DALY) to changes in construction (insulation, air tightness) and building services engineering (ventilation, heating) for dwellings under Dutch conditions. The most important conclusion is that natural ventilation, mechanical ventilation and balanced ventilation not only differ in their average health effect, but even more so in the size of the ranges of these effects. Other systems, such as heat pumps or photo voltaic cells are expensive but will become economically applicable when healthcare costs are taken into account. These outcomes gave valuable clues for product innovation and opened the possibility to model health in relation to built environments. The method could also be applied to quality classification systems for dwellings.
Jan L.M. Hensen
Center for Buildings & Systems TNO – Eindhoven University of Technology, The Netherlands

keywords: Building Performance Simulation, Design Support
abstract: This paper discusses some issues which hinder effective use of building performance simulation in building design, and some approaches towards better and more efficient use of this important but underutilized technology. In particular, the paper discusses the issues of quality assurance, the relative slow software developments and the limited use (usability) of building performance simulation mainly during the final stages of the building design process.
9 Agent Technology 2
Linda Ma, Theo Arentze, Aloys Borgers, and Harry Timmermans
Eindhoven University of Technology, The Netherlands
keywords: Land Suitability Analysis, Multi-Agents, Knowledge Representation, Bayesian Decision Networks.
abstract: Land suitability analysis typically involves the assessment of the suitability of land units without knowing the future spatial distribution of land use. Traditional planning techniques have used “algebraic equations” to express land suitability as a weighted function of suitability scores across multiple criteria. However, the existing multi-criteria evaluation methods do not systematically account for uncertainty about the land use in adjacent and other cells. This paper proposes an alternative approach to land suitability analysis that does address the problem of uncertainty. In particular, Bayesian decision networks are suggested as a means of knowledge representation for agents in a multi-agent land use simulation system. Bayesian decision networks model the uncertainty in terms of probabilities specified in the network representing the expertise of specialists with respect to specific land uses. This paper discusses the approach and illustrates its use in the context of a retail agent.
Jan Dijkstra, Harry J.P. Timmermans, and Bauke de Vries
Eindhoven University of Technology
keywords: Multi-Agent Systems, Activity Modelling, Pedestrian Behaviour
abstract: The Amanda model system uses a multi-agent approach in an attempt to simulate pedestrian dynamic destination and route choice. Data collecting efforts are needed to calibrate the model. This paper discusses these data requirements.
Romulo Krafta
Rio Grande do Sul Univerity – Urban & Regional Planning Unit – Brazil
keywords: Urban Morphology, Urban Growth, Simulation
abstract: The purpose of this paper is to report efforts towards the construction of a model for urban spatial dynamics simulation, based on multi-agents and space. The underlying idea is to have urban space producers and consumers operating in a two-layer, two-circuit model. The first layer holds urban space and its successive transformations; a second layer contains agents related to space; the first circuit simulates space production, and a second one simulates space consumption. Relationship between layers is represented as objective spatial features that agents are submitted to (the machine) and subjective meanings agents attach to each spatial feature (the ghost). While space works always in the same way, meanings vary according to each agent’s background and context. Relationships between circuits are represented by means of a market game in which producers try to maximize their profits by gambling with their risks, whereas consumers try to foresee the spatial distribution of local externalities that maximizes their utilities and investments. Urban Spatial Features are captured through centrality and land use patterns; every single agent’s action leads to changes in both patterns. Producers’ profit is a function of built form location. Consumers’ local externalities are concerned basically with present and future services. The model iteration is twofold: first it generates and allocates a number of built forms within a previously determined spatial system (a cellular matrix, for example), and second it allocates users to built forms. Population of users have its social profile and growth rate externally determined. Built form allocation is decided on the basis of a combination of profitXrisk perspectives. Users’ locational choice is supported by accessibility to services and present/future neighbourhood profile. Built form allocation works as parameter for users’ locational assessment, whereas users’ choices are used as parameters for developers. The model tends to adjust itself, in terms of quantities and types of built forms to be erected, although through a market lag of some iterations. Allocations are always made through weighted draws, so that mutations (non deterministic allocations) do occur.
10 Collaborative Design
A. Penn, C. Mottram, A. Fatah gen. Schieck, M. Wittkämper, M. Störring, O. Romell, A. Strothmann, and F. Aish
The Bartlett Graduate School, University College London,UK; Fraunhofer FIT; Aalborg University; SaabAvionics, Linie4 Architekten; Foster and Partners

keywords: Design Collaboration, Tangible Interface, Gesture, Agent Simulation, Augmented Reality.
abstract: Immersive virtual environments have received widespread attention as providing possible replacements for the media and systems that designers traditionally use, as well as, more generally, in providing support for collaborative work. Relatively little attention has been given to date however to the problem of how to merge immersive virtual environments into real world work settings, and so to add to the media at the disposal of the designer and the design team, rather than to replace it. In this paper we report on a research project in which optical see-through augmented reality displays have been developed together with prototype decision support software for architectural and urban design. We suggest that a critical characteristic of multi user augmented reality is its ability to generate visualisations from a first person perspective in which the scale of rendition of the design model follows many of the conventions that designers are used to. Different scales of model appear to allow designers to focus on different aspects of the design under consideration. Augmenting the scene with simulations of pedestrian movement appears to assist both in scale recognition, and in moving from a first person to a third person understanding of the design. This research project is funded by the European Commission IST program (IST-2000-28559).
Sheng-Cheng Shih
Graduate Institute of Architecture, National Chiao Tung University, Hsinchu, Taiwan
keywords: Interoperable, Co-Design Support, Distributed Hybrid Agent Architecture, Collaborative Design-Build
abstract: The interoperable model of Co-design process and systems characteristics that is required to reduce the adversarial nature of most construction projects. Interoperable Co-Design System was used successfully to complete the Si-soft Research Center of Taiwan. This project-alliancing project was analyzed as a case study to evaluate the validity of the system. Main paradigms of the Co-Design were reviewed and numerous examples from the management of this project were cited that support the theoretical recommendations of this model. It was concluded that the system use wild client/server network architecture embedded with peer-to-peer agent technology to provide an open, familiar and easily extended co-design system.
N. Deshpande, B. de Vries, and J.P. van Leeuwen
Eindhoven University of Technology, The Netherlands
keywords: Collocated Multi-Disciplinary Collaborative Design (CMCD), Shared Understanding, Computer Supported Collaborative Work (CSCW), Group Activity, Design Space, Groupware
abstract: This review identifies the research findings in the area of computer mediated, collocated, multi-disciplinary collaborative design. After a general look into this and related research areas, three influencing aspects of collaborative design are reviewed. Design activities, working together, and collaborative systems that fit into various work practices. Finally, we note missing points of research in the area of collocated, multi-disciplinary, collaborative design.
Jakob Beetz, Jos van Leeuwen, and Bauke de Vries
Eindhoven University of Technology, The Netherlands
keywords: Multi Agent Systems, Design & Decision Support Systems, Collaborative Design, Human Computer Interfaces, Machine learning, Data Mining
abstract: In this paper we are drafting the outline of a framework for a Multi Agent System (MAS) for the support of Collaborative Design in the architectural domain. The system we are proposing makes use of Machine Learning (ML) techniques to infer personalized knowledge from observing a users’ action in a generic working environment using standard tools such as CAD packages. We introduce and discuss possible strategies to combine Concept Modelling (CM)-based approaches using existing ontologies with statistical analysis of action sequences within a domain specific application. In a later step, Agent technologies will be used to gather additional related information from external resources such as examples of similar problems on the users hard disk, from corresponding work of team-members within an intranet or from advises of expert from different knowledge domains, themselves represented by agents. As users deny or reward resulting proposals offered by the agent(s) through an interface the system will be enhanced over time using methods like Reinforced Learning.


Book References

Page-numbers refer to the two books of proceedings as follows:

RA = Van Leeuwen, J.P. and H.J.P. Timmermans (eds.) 2004,
Recent Advances in Design & Decision Support Systems in Architecture and Urban Planning,
Dordrecht: Kluwer Academic Publishers. ISBN 1-4020-2408-8.
D = Van Leeuwen, J.P. and H.J.P. Timmermans (eds.) 2004,
Developments in Design & Decision Support Systems in Architecture and Urban Planning,
Eindhoven: Eindhoven University of Technology. ISBN 90-6814-155-4.