Modeling joint activity-travel patterns in pedestrian networks with use of Wi-Fi data

Abstract

This paper proposes a new activity-based equilibrium model to solve a joint activity-travel pattern scheduling problem in pedestrian networks. A novel supernetwork platform, pedestrian-joint-activity-time-space, is proposed to capture the complex structure of pedestrian networks and intragroup interactions on joint activity choices of pedestrians. The joint activity-travel pattern scheduling problem is transformed into an extended static user equilibrium traffic assignment model with side constraints for group interactions. A heuristic solution method is proposed to solve the user equilibrium problem concerned. A simplified pedestrian network is used to illustrate the convergence of the proposed heuristic solution method. Furthermore, the use of Wi-Fi tracking data from mobile phones to generate the activity patterns of pedestrians and their interactions is also investigated with a real case study in a selected area of The Hong Kong Polytechnic University.