Network-wide Traffic Signal Optimization under Connected Vehicles Environment

Connected vehicle technology holds great promises for next generation of traffic signal control. This paper proposes a novel mixed-integer linear programming (MILP) model to optimally coordinate traffic signals at the network level under connected vehicles environment. We leveraged the combinatorial Benders decomposition method to solve this large-scaled MILP model more efficiently than the SCIP solver. Numerical experiments show that our approach can significantly improve network traffic operations compared to a benchmark signal plan in terms of vehicle delay, travel time and throughput.

I. Introduction

Connected vehicle (CV) is a technology that enables communications among vehicles (V2V), vehicles and infrastructure (V2I), e.g., traffic signals and road signs, and even vehicles and pedestrians (V2P). Data that record the states of traffic participants can be transmitted and exchanged through Dedicated Short Range Communications (DSRC), Cellular Vehicle-to-Everything (C-V2X), and Bluetooth. In addition, the emerging 5G technology that has much faster speed, wider transmission range and lower latency is expected to boost connected vehicle technology and its wider applications [1].