Mixed Traffic Flow Signal Timing Optimization Method Considering E-Bike Expansion Influence

Abstract

The increase in e-bikes has aggravated traffic conflicts and casualties at intersections. However, existing signal control methods focus on motor vehicles, ignoring the influence of expansion behavior of e-bikes on motor vehicles. In order to better balance the traffic benefits between them, this paper proposes a new release mode named e-bike early green and establishes a signal timing model of mixed traffic flow considering e-bike early green (MTEG). In particular, an effect strength indictor that reflects the degree of influence of e-bikes on motor vehicles is put forward. Then, based on this indicator and e-bike ratio, the calculation model for e-bike early green time is built. The Nondominated Sorting Genetic Algorithm II is applied to solve the MTEG model that considers multi-objective optimization coordination. The results show that, when the e-bike ratio is in the range of 0.4–0.6, compared with the method proposed by the Transport and Road Research Laboratory (TRRL), the method proposed by the Australian Road Research Board (ARRB), and Improved-Webster method, the maximum improvement values of the MTEG method in cycle length, average vehicle delay, intersection capacity, and average stops per vehicle are −5.56%, −22.04%, +2.30%, and −8.00%, respectively. The outcomes provide a signal timing basis and technical support for a mixed traffic flow environment containing numerous e-bikes.