Can Artificial Intelligence warn of overcrowded platforms?
The crowds on platforms are increasing. But when is full, too full? And if overcrowding is imminent, is there a way to warn railway personnel in advance? In order to effectively operate a station in critical situations, the current situation must be closely monitored and future scenarios reliably predicted. With ASE's Pedestrian Analytics System (PAS) it is possible to accurately measure passenger flows and densities to gain valuable information on station usage and predict potentially critical situations.
Jessica Weibel (ETH Zurich) has written her master thesis concerning this topic in collaboration with ASE. Her thesis deals with machine learning algorithms that can be used to optimize the predictions regarding the utilization of platforms at Amsterdam Zuid station. Want to know more? Contact us now.