摘要
为应对传统动态需求下单一车辆配送成本高、时效差等问题,通过共享地铁剩余运能将地铁网络融入城市末端配送网,提出基于地铁和货车联运的动态选点-路径问题。从联合运输配送成本最小化角度出发,考虑动态需求、客户服务时间窗、多转运点接送等因素,构建地铁转运点选择、货车调度以及接运配送路径的两阶段动态选点-路径整体优化模型。针对模型特征,设计“初始阶段+动态阶段”的两阶段启发式算法,将Mark-Sweep算法和蚁群算法相结合求初始解,再设计双层启发式集成算法求动态解,为提高算法的求解质量和效率,外层遗传算法采用矩阵编码结构并通过截取掩码方式改进交叉变异算子,内层蚁群算法采用关键值编码结构并改进多点接送概率选择操作,通过内外层信息交互实现选点和路径的集成优化。最后结合不同算例验证模型与算法的合理性及有效性。结果表明,基于地铁和货车联合运输开展动态配送,不仅能有效提高配送时效性和精准度,而且能使配送成本降低约18%。
To deal with the high cost and poor timeliness of the single transportation mode in urban terminal vehicle distribution under the traditional dynamic demand,integrating subway network into the urban terminal vehicle distribution system by sharing the surplus capacity of subway to transport freight is an effective way.The overall optimization model of two-stage dynamic point-routing problem for subway transfer point selection,truck scheduling and pickup and distribution routes was established to minimize distribution cost,considering the factors of dynamic demand,customer service time window,multi-transit point pick-up and delivery.According to the characteristics of the model,a two-stage heuristic algorithm including initial stage and real-time stage was designed to find the initial solution combining the Mark-sweep algorithm and ant colony algorithm.A bi-level heuristic integration algorithm with matrix coding structure plus the key-value coding structure was designed to find the real-time solution.To improve the solution quality and efficiency,the outer tier genetic algorithm with matrix coding structure improved the crossover mutation operators by intercepting masks.The inner tier ant colony algorithm with key value coding structure improved the multi-point pickup and delivery probability selection operation.The integrated optimization of point selection and routing was realized through the interaction of inner and outer tier information.Different examples were designed to verify the rationality and effectiveness of the model and algorithm.The results show that the timeliness and accuracy of the intermodal transportation model in this paper are better than the single transportation model,with a reduction of 18%of the distribution cost.
作者
崔瑶
周晓晔
何亮
CUI Yao;ZHOU Xiaoye;HE Liang(School of Management,Shenyang University of Technology,Shenyang 110000,China;School of Management,Liaoning Institute of Science and Technology,Benxi 117000,China;Department of Mechanical and Electronic Engineering,Liaoning Provincial College of Communications,Shenyang 110000,China)
出处
《铁道学报》
EI
CAS
CSCD
北大核心
2023年第1期9-19,共11页
Journal of the China Railway Society
基金
辽宁省社会科学规划基金(L22ZD010)。
关键词
地铁
联合运输
动态配送
选点-路径问题
双层启发式集成算法
subway
intermodal transportation
dynamic distribution
point-routing problem
bi-level heuristic integration algorithm
