Experimental Investigation on Information Provision Methods and Guidance Strategies for Crowd Control

摘要

Recent improvements in crowd sensing and dynamic signage are paving the way for automatic crowd control, in which real-time information is used to steer crowds of people. However, little is known on the impact that automatic information provision has on crowd dynamics. In this study, pedestrians moving in a two-lane fork loop were given different types of information and were guided using different strategies. Our analysis focused on which combination had the best performance and was perceived as comfortable from the participants. Results indicate that a human guidance is preferred when comfort is concerned, but providing real-time traffic information is more efficient in terms of pedestrian flow. It is speculated that the processing of a large volume of information limits the perceived comfort when realtime data are provided. As a consequence, people prefer the less effective human guidance in which decision-making is entrusted into others. With this said, the speculative nature of this conclusion puts emphasis on future studies which will need to investigate more in detail how information provision affects pedestrians' behavior and comfort perception in public spaces.