Quantitive description of jam formation in pedestrian streams

Laboratory experiments in schools

First laboratory experiments were executed in November 2014 with pupils of grade 5 and 11.

Awareness for pupils of the dangers in large crowds or during the evacuation of buildings.

Research of the influence of the heterogeneity of pedestrians under the parameter of body size on pedestrian dynamics. Special focus is the study of congestion in a crowd, as well as the safety margins/distance and the walking speed of single person in this crowd.

Influence of groups on the pattern of movement of pedestrians. This applies in particular to clarify the extent to which the formation of pedestrian groups affects the dynamics of large crowds and thus on the evacuation time of rooms and buildings. We also investigate in this context whether a conductive lead person in the group has a positive effect on the evacuation process.

More experiments will follow in spring of year 2015.

Objectives

For the design of escape routes in buildings and large scale events reliable specifications and tools for the dimensioning of pedestrian facilities are needed. However, up to now fundamental questions regarding the dynamics of large pedestrian streams are open. This is reflected in conflicting building regulations within Germany. The aim of our project is to achieve a reliable quantitative description of pedestrian streams by experiments and modeling.

Results from models indicate that the heterogeneity of pedestrians influence the formation of jams in pedestrian streams. In the proposed project this hypothesis will be tested experimentally. We expect to improve our understanding of the emergence of critical situations in pedestrian streams. In the modeling we will try to improve on recent developments where the focus was on the precise modeling of volume exclusion. Now we try to realize an improved steering of the agents, considering a forecast of the environment. With this we want to describe quantitatively the dynamics of pedestrian streams at corners, T-junctions and bidirectional streams.

We are working on this project together with Prof. Andreas Schadschneider, Institute for Theoretical Physics, University of Cologne.