A Subway Station Platform Crowd Simulation System

Abstract

A crowd simulation system in a subway station environment is proposed to address the problem that most of the platform passengers in the existing subway simulation systems will use fixed routes which has been planned before to get on and off the train. Along with the creation of the basic model of crowd behavior in the subway station, passengers will obtain the navigation grid and door information through the environment information generator. And, through a finite state machine and the usage of the A* algorithm which can achieve wayfinding, passengers can manage their own state in the intelligent passenger system. When selecting a screen door, passengers will use the algorithms of Topsis analysis and fitting formulas to simulate real decisions. Finally, the simulation-based passenger behavior is compared with the real situation to verify the correctness of the system.