This approach differs of the classic four stage planning model. The proposed model incorporates the distribution and assignment stages in an interlaced form, with a dynamic behaviour along a specific time horizon. Therefore it does not characterize the mobility demand like origin-destination trip matrix. In each instant, the pedestrians (entities) try to minimize their travel time toward destinations under the user's equilibrium hypothesis, an equilibrium situation that is impossible to reach, due to the dynamic and continuous character of the process. The model enables the study of important expositive events (fairs and international expositions). This paper shows the application of this model to the International Exposition of Zaragoza 2008, simulating one design day, with an attendance of almost 100,000 people in 17 hours.

The model can be applied in a non-pedestrian context, as the basis for the study of phenomena governed by the attraction of vehicles toward specific zones, such as parking in central areas of cities.

During the 1960s and 1970s a handful of pedestrian demand models were developed with the objective to forecast pedestrian flows and prioritize pedestrian improvements in central business district areas. These models were developed with a structure similar to the four-stage transportation planning models, including trip generation based on land use characteristics, trip distribution and assignment over a network following a user equilibrium model approach.

Related with this paper, Ness et al. [1] used a gravitational model to predict pedestrian flow volumes in Toronto central business area. The business district was divided into different zones and a pedestrian network was defined using pedestrian links, street junctions and zone centroids. Numerous generation and attraction measures for office areas and public transportation lines were take into account. The method considered speed-density factors and a set of minimum paths were defined as input data for a gravitational distribution model. Shortest path calibration was made using travel time data, delays at junctions, attraction measures for main streets and penalty turn functions.

That is the basis of the DDAS model presented in this paper. The method does not follow a user equilibrium situation, rather a succession of equilibrium states inside a dynamic system that evolve during a complete journey. The building crowding-in effect over people change during the day as a function of events programmed beforehand. The perception of pedestrians about the different paths towards their destinations differs for each subject. The model follows a stochastic approach to assign routes to pedestrians.

The aim of this model is not only the analysis of the transport system inside the scenario (considering mechanized links, escalators, conveyors, ride vehicles, bicycles, electrical cars, monorail, etc) but also the capacity study of different kinds of links, squares, footbridges and the evaluation of queue sizes, visit times and their evolution.

1

Ness, M.P., J.F. Morrall, and B.G. Hutchinson. An Analysis of Central Business District Pedestrian Circulation Patterns. Highway Research Record 283, 1969.

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £105 +P&P)