考虑时间窗的通勤定制公交线路优化

Optimization of customized bus routes for commuting considering time windows

针对常规定制公交线路规划方法不能满足乘客高峰通勤需求等问题,在考虑乘客期望时间窗和最大容忍时间窗的基础上,引入柔性时间窗的概念,基于通勤者居住地与工作地时空分布差异性特点分别设置上下车站点罚函数,以车辆固定成本、车辆运行成本、车辆延误惩罚成本和未提供乘车服务惩罚成本之和最小为优化目标,以时间窗、运行距离和最大站点数量等为约束,建立考虑柔性时间窗的通勤定制公交线路优化模型。在兰州市通勤乘客数据基础上,选取两组大小不同规模的算例集,使用Gurobi和传统TS算法求解小规模算例与基于目标控制原则的动态禁忌长度TS算法求解大规模算例验证模型有效性。结果表明,在小规模算例中Gurobi求解可得到较优的目标值,但随着数据规模的增加其求解效率逐渐劣于TS算法,设置柔性时间窗车辆提供乘车服务的人数比硬时间窗背景下增加了276人,且有效规避了软时间窗背景下车辆到达末站点的时刻波动。本文所提方法在均衡乘客和运营方利益的同时,可通过调节柔性系数产生多元化的运行方案,能够为通勤定制公交运营及管理提供决策支持。

Conventional customized bus route planning methods cannot meet the peak commuting demand of passengers.Therefore,to address this shortcoming,the concept of a flexible time window is introduced based on passenger expectation time window and maximum tolerance time window.The penalty function of boarding and alighting stations is set separately based on the spatial and temporal distribution differences between commuters’ residences and workplaces.The sum of vehicle fixed cost,vehicle operation cost,and vehicle delay penalty cost and failure to provide ride service are set as the optimization objectives,whilst the flexible time window,operating distance,and maximum number of stops are used as constraints to establish the optimization model of customized commuter bus routes considering the flexible time window.Based on the data of commuter passengers in Lanzhou City,two small-scale and large-scale cases were selected,and the proposed model was verified using Gurobi and the traditional TS algorithm to solve the small-scale cases and the dynamic taboo length TS algorithm based on the objective control principle to solve the large-scale cases.The results show that,as the data size increases,the Gurobi solution can obtain better target values in small-scale cases,however,its solution efficiency becomes progressively inferior to that of the TS algorithm.In addition,the number of rides provided by vehicles in the flexible time window increased by 276 people compared with that in the hard time window background,and the fluctuation of the vehicle arrival time at the end station in the soft time window background was effectively avoided.The method proposed in this paper can balance the interests of passengers and operators,whilst simultaneously generating diversified operation schemes by adjusting the flexibility factor,which can provide decision support for customized commuter bus operation and management.