Environment and Planning B: Urban Analytics and City Science

Vulnerable road users have steadily attracted increased importance in transport and planning. The behaviour of pedestrian movements (especially in the areas off but adjacent to roads) requires improved tools to address the issues now being raised. Such behaviour and interactions can now be modelled by using a combination of massively parallel processes simulating individual pedestrians, and a series of behaviours of these simulated pedestrians in the interactions with each other and their environment. The PEDFLOW model has been implemented in the parallel processing language Occam as an agent-based evolutionary system, which allows extensive modelling of detailed pedestrian behaviour with minimal complication. The principles and methodology of its development and application are specified.

Pedestrian movement models have been developed since the 1970s. A review of the literature shows that such models have been developed to explain and predict macro, meso, and micro movement patterns. However, recent developments in modelling techniques, and especially advances in agent-based simulation, open up the possibility of developing integrative and complex models which use existing models as ‘building blocks’. In this paper we describe such integrative, modular approach to simulating pedestrian movement behaviour. The STREETS model, developed by using Swarm and GIS, is an agent-based model that focuses on the simulation of the behavioural aspects of pedestrian movement. The modular structure of the simulation is described in detail. This is followed by a discussion of the lessons learned from the development of STREETS, especially the advantages of adopting a modular approach and other aspects of using the agent-based paradigm for modelling.

Although pedestrians have individual preferences, aims, and destinations, the dynamics of pedestrian crowds is surprisingly predictable. Pedestrians can move freely only at small pedestrian densities. Otherwise their motion is affected by repulsive interactions with other pedestrians, giving rise to self-organization phenomena. Examples of the resulting patterns of motion are separate lanes of uniform walking direction in crowds of oppositely moving pedestrians or oscillations of the passing direction at bottlenecks. If pedestrians leave footprints on deformable ground (for example, in green spaces such as public parks) this additionally causes attractive interactions which are mediated by modifications of their environment. In such cases, systems of pedestrian trails will evolve over time. The corresponding computer simulations are a valuable tool for developing optimized pedestrian facilities and way systems.

In this paper I briefly review the use of computer simulation in archaeology and argue that pedestrian modelling has the potential to overcome many of the problems associated with earlier simulation studies. I then introduce the MAGICAL simulation software, which was written to facilitate the use of multiagent simulation within a geographical information system. In the final part of the paper I describe the use of MAGICAL to study the evolution of cultural learning among early hominids.

Our aim in this paper is to build and test a model which classifies and identifies pedestrian shopping behaviour in a shopping centre by using temporal and spatial choice heuristics. In particular, the temporal local-distance-minimising, total-distance-minimising, and global-distance-minimising heuristic choice rules and spatial nearest-destination-oriented, farthest-destination-oriented, and intermediate-destination-oriented choice rules are combined to classify and identify the stop sequences and route choices of shopping pedestrians. First, several linear networks with a single entry node and a few stop nodes are investigated. For these networks, the global-distance-minimising and spatial choice heuristics classify and identify the sequences of stops very well. Although the local-distance-minimising choice rule identifies pedestrian route choice quite well, another heuristic is needed to improve the identification. In this paper a new, attractive-street-oriented heuristic is suggested to improve the identification ability of the model. This choice rule suggests that shopping pedestrians will never leave the attractive shopping streets before completing their shopping. The model is then applied to empirical data of pedestrian shopping behaviour in Veldhoven City Centre in The Netherlands. The findings of this application suggest that the model based on choice heuristics might be useful to classify and identify the sequences of stops and route choice behaviour of shopping pedestrians in a shopping centre.

Local governments rarely use risk analysis, the most sophisticated level of hazard assessment, to inform their planning and development decisions. But new tools are rapidly becoming available to accomplish such analysis. In this paper we present an example of an earthquake risk analysis for Los Angeles County, using available land-use maps, a probabilistic earthquake hazard model developed by the Southern California Earthquake Center, and the Federal Emergency Management Agency's new HAZUS earthquake loss estimation software. We computed the annual expected loss owing to earthquakes and the spatial variation of this risk. The analysis shows that the annual long-term earthquake risk in Los Angeles County, as a result of direct structural and nonstructural damage, is $388 million per year. We also investigated the extent to which planned future land-use growth would affect this risk estimate. We found that planned growth of 14.2% would result in an increase in annual risk to $449.5 million, a 15.8% increase over the risk to current land uses. Because of ever increasing disaster costs, planners need to be able to evaluate the risks that their communities face, both in the present and in the future. It is particularly important for planners to be sure that they are not disproportionately planning future growth for hazardous locations. In this paper we describe some ways in which to perform such evaluations, by using tools that have only recently become available.

It has been recognized that urban residential segregation is a scale-dependent phenomenon, and yet all existing structural or spatial indices measure only certain dimensions of segregation at one single scale. Inspired by new metrics developed by landscape ecologists to measure landscape heterogeneity, in this paper we explore the feasiblity of a multiscale, lacunarity-based segregation measure. We also develop a straightforward GIS-based procedure to calculate this new measure. Our initial simulation results show that lacunarity is an effective measure that can capture multiple dimensions of segregation patterns at multiple scales.

Urban evolution is composed of two interlinked phenomena. Over time, changes occur in urban size as measured, for example, by population. The geographic space occupied by human activities and by buildings also changes over time. Those two aspects of urban evolution are linked and, as such, should be studied in tandem. In this paper we present an analysis of the process of formation of urban centers as a result of growth spurts and the structured behavior of individuals. We propose a dynamic model of the spatial self-organization of a town. The model yields descriptions of the temporal evolution of the urban population and of the space occupied by it. At the backdrop of the model is a particular conception of the behavior of land developers. The evolution of the town emerges from the juxtaposition of centripetal and centrifugal forces that represent the key elements in the developers' behavior. Through computer simulations, the model has been applied to Petah Tikvah, a small town in the Tel Aviv metropolis. The comparison between the simulations and the real town growth is very good.

In this paper I explore values in conserving the built environment as applied by local planning authorities in two towns of different historical character, although comparable in size and statutory conservation responsibilities. A two-tier investigation involved exploring the conservation practices of these local authorities and examining four recent development cases. Interviews with a wide range of local players built up a picture of the perceptions of local conservation culture. The results suggest that tensions are created as the national emphasis towards conservation's more flexible contribution to regeneration is exposing weaknesses in some underlying assumptions. In drawing away from the traditional development control aspect and covering such a breadth of justifications, this creates professional, political, and public misunderstandings which may undermine support for conservation.