Implementation of an Automatic Control Strategy to Minimize Headway Variance

Abstract: In this project, a model of a high-demand bus line is formulated and simulated. Three different control strategies, Timetable, Headway Thresholding and Self-Coordinating, are implemented and tested in the simulation, in order to establish one that reduces passenger wait time and that resists and recovers from bunching. Two of the three strategies uses inter-vehicle headways to control the flow of the buses, requiring real-time information sharing of vehicle positions between the buses. With the aid of so-called Automatic Vehicle Location-systems (AVL) already in use in many public transit systems, this information could be shared instantly through the Cloud and be used to make headway-based control strategies a reality. An interactive graphical user interface supplements the simulation software. This can be used as an educational tool for teaching about control theory and the dynamics of public transit systems.