考虑到公交客流时间分布非均衡以及公交区域调度的多条线路客流高峰时间、空间分布集中会导致总体客流的高低峰客流量差异进一步加大，提出在低平峰期允许部分公交车辆停运的抽停策略，从“部分车次被某辆车完成”的集合划分角度研究区域公交车辆调度问题，建立改进的区域公交车辆调度优化模型，并设计了最大最小蚁群算法进行求解． 算例结果显示: 与常规调度方案相比，车辆等待时间减少4034min，等待时间在总工作时间中占比减少10%，驾驶员减少14 个，证明了文中模型的有效性和实用性．

The distribution of the bus passenger flow in time is uneven,and in the multiple lines with the regionalbus scheduling,the concentrated distribution of passenger flow peaks in time and space can make a bigger passengerflow difference between peak and off-peak hours.In order to solve these problems,this paper proposes a removingand parking strategy which allows some buses to stop running in off-peak hours.Then,by analyzing the regionalvehicle scheduling problem from the viewpoint of set partitioning which means that a vehicle finishes some trips,an improved optimization model is built for the regional bus scheduling,and a MMAS( Max-Min Ant System) algorithmis revised to solve this model.Case analysis shows that,as compared with the conventional schedulingscheme,the proposed model shortens the bus waiting time by 4034 minutes,reduces the rate of bus waiting time inthe total working time by 10% and decreases the driver number by fourteen,which proves that the proposed modelis effective and practical.