A holding strategy to resist bus bunching with dynamic target headway. (February 2020)

• Record Type: Journal Article
• Authors: He, Sheng-Xue; Liang, Shi-Dong; Dong, June; Zhang, Ding; He, Jian-Jia; Yuan, Peng-Cheng
• Is Part Of: Computers & industrial engineering. Volume 140(2020)
• Publication Date: 2020-02
• Subjects: Public transit -- Bus bunching -- Stability -- Adaptive control -- Holding strategy; Engineering -- Data processing -- Periodicals; Industrial engineering -- Periodicals; 620.00285
• DOI: 10.1016/j.cie.2019.106237
• Languages: English
• ISSNs: 0360-8352

Abstract

Highlights: New holding strategy with dynamic target headway is proposed. Convergence under multiple control points is proven theoretically. Enlarged set of control points may deduce the stability of bus line. Too many control points will increase the average onboard riding time. High stochastic level of travel time will lower the effect of stabilizing. Abstract: Bus bunching commonly appears in a high-frequency bus line and will bring the lengthened waiting and riding times to passengers. Among the existing strategies of resisting bus bunching, only a few tried to cope with a high-frequency bus line which has not any pre-specified schedule or headway required to be adhered to. To stabilize a bus line as above, we propose a target-headway-based holding strategy in this paper. The average of the instantaneous time headways is used as the dynamic target headway. Based on the difference between the dynamic target headway and the forward headway of the current bus, the new strategy determines a proper holding time in a dynamic traffic situation. After removing the stochastic factors influencing on the headway, we proved the effectiveness of the new strategy to stabilize an unstable bus line. Different from the existing theoretical analyses which only focus on the single control point situation, the theoretical analysis in this paper takes into account any given number of control points. The numerical experiment demonstrated the effectiveness of the strategy by comparing with other strategies. The following insights were uncovered from the numerical analysis. With the increase of control points in a bus line, the stability performance and the service level of the bus line will increase at the beginning and then change to decrease. With the increase of the stochastic level of travel time, the performance of the bus line will decrease gradually. The new strategy can effectively cope with the non-peak, single-peak and double-peak demand patterns in a similar way.