考虑不确定电量消耗和充电站排队的纯电动车城市物流配送路径优化

While air pollution being a major issue in the country, the government is paying more attention to the usage of green vehicle. Given the incentives from government, delivering goods via electric trucks in cities becomes a trend in logistics practices. The routing and charging plans of electric vehicles depend on several factors that will dynamically change along the trip. However, previous studies in this area are limited and only focus on deterministic electric vehicle routing problems. In this project, we propose to study a stochastic electric vehicle routing problem with multiple vehicles, time windows and limited vehicle freight capacities. The consumption rate of energy is uncertain and affected by several factors, and the electric vehicle may need to be recharged on route. We address the problem from the following aspects. First, we consider that the battery level of the vehicle is affected by several factors (such as the temperature, vehicle load, and how the vehicle is being driven) and will decrease stochastically. Second, we consider the situation that the vehicle may need to wait in a queue to get charged at a station. Because the charging time of other vehicles in the queue depends on their remaining battery level, which is uncertain, the waiting time for the vehicle in the queue is uncertain. Third, we take into account the traffic condition while routing the electric vehicles. We model the above problems via stochastic programming and dynamic programming models. We focus on developing heuristics methods as well as approximate dynamic programming algorithms to solve the problems. To test the effective of our modeling and solution approaches, we run computational experiments and conduct field studies in collaborative logistics corporations.

随着我国空气污染问题的日益凸显,国家对新能源汽车的重视程度不断提高。在相关政策的支持下，将纯电动轻卡汽车用于城市配送成为物流行业发展的新趋势。本项目研究纯电动车在城市物流配送中的路径优化问题，以多车辆、带顾客时间窗、受车载容量限制、配送中途需要补电的路径问题为基础，考虑纯电动车在配送过程中的电量消耗受多种不确定因素影响、充电站排队、以及动态变化的交通网络状况。首先研究当纯电动车电量消耗受车辆行驶速度、环境温度、载重量、道路状况等多种因素影响，电池电量呈非线性变化时的物流配送路径优化；然后在此基础上考虑纯电动车需要在充电站排队充电，且排队等待时间不确定的情况；最后加入对实时交通网络状况的考虑。本项目拟对以上问题建立随机规划模型和动态规划模型，设计启发式算法和近似动态规划算法求解，通过计算机仿真来验证算法的有效性，进行案例分析，将研究结果应用于企业的实际运营中。

本项目研究考虑不确定电量消耗和充电站排队的纯电动车城市物流配送问题。在具体实施过程中，主要完成以下问题的研究：1）电量消耗受不确定因素影响的纯电动车路径优化问题；2）充电站排队且排队时间不确定的纯电动车路径优化问题；3）考虑交通拥堵收费的多车型车辆路径优化问题；4）车辆时空轨迹数据的挖掘问题。项目团队为这些问题建立了具有针对性的随机规划模型和混合整数规划模型，并基于模型设计了高效的求解算法，包括近似动态规划算法、混合滚动近似动态规划算法、两阶段启发式算法、基于禁忌搜索的LANCOST算法等。我们通过仿真实验展示了所设计算法的优异性能。项目期间共发表（含录用）6篇SCI/SSCI索引论文，其中部分论文发表在知名的国际期刊上，如《Transportation Science》、《European Journal of Operational Research》、《Networks》等。项目组共有5人次参加国际顶级会议并作报告，共培养博士生3名。