A cell.based propagation model for pedestrian flows in railways stations

Abstract

Pedestrian flows occurring in train stations are by nature multidirectional and highly non-stationary. In this work, we present a cell-based pedestrian flow model capable of describing flow patterns arising when a multitude of trains arrive and depart in close succession. Based on first-order pedestrian flow theory and a cell-transmission model, characteristic flow patterns and local densities can be realistically reproduced. Furthermore, individual walking times can be predicted for groups of pedestrians depending on route, departure time and walking-specific attributes (e.g. ‘in a hurry’). A detailed derivation of the mathematical framework and of the computational architecture as well a literature review is provided. A real case study is discussed, underlining the applicability of the proposed model.