Identification of Split Factors in Pedestrian Flows Modeling

The paper is devoted to mathematical modeling of indoor pedestrian flows. The model under consideration is an analog of the cell transmission macromodel (CTM). The possibility of identifying split factors that were previously assumed to be a priori given is studied. These factors denote the proportions in which a flow of people is divided when moving to other rooms from the current one. We propose an identification algorithm based on interval estimates of both forward and backward reachability sets. The algorithm is illustrated with a numerical example.