A multi-stage stochastic optimization approach to the stop-skipping and bus lane reservation schemes

Abstract

The bus stop-skipping scheme is designed to reduce the number of stops and increase bus operating speed. But in a mixed traffic environment, buses can hardly maintain a high and stable operating speed due to frequent merging and overtaking activities. In this regard, the exclusive bus lane is used to separate buses from normal traffic. However, the permanent bus lane would inevitably decrease road capacity and result in more congestions along the bus lane, especially during peak hours. This paper proposes a hybrid bus operational scheme combining the bus lane reservation strategy and the stop-skipping control. Considering the uncertainty of passenger demand, this paper proposes a multi-stage stochastic programming model. The evolution of the uncertainty is interpreted by a scenario tree. A progressive hedging-based method is developed to separate the large-scale optimization problem into sub-problems. Two numerical experiments are conducted to verify the proposed model and solution algorithm.