论文部分内容阅读
针对我国城市轨道交通(以下简称城轨)乘务计划编制效率较低的现状,结合城轨乘务劳动作业规定,建立城轨乘务任务划分的RTSCP模型。提出基于列生成思想的乘务任务划分优化算法(CGLR算法),采用该思想获取小规模较优乘务任务子集合,降低任务划分问题的求解复杂度;采用以最优拉格朗日乘子为启发信息的LR_Heuristic算法取代单纯形算法求解RTSCP松弛问题,提高算法效率;结合获取的拉格朗日乘子,引入随机列修补技术获取RTSCP问题的可行解,提高解质量。最后以某地铁线路为背景进行验证。结果表明,模型及算法能有效求解乘务任务划分问题并获得较优的划分方案。
Aiming at the current situation that the planning of the flight planning of urban rail transit in our country is relatively inefficient, the RTSCP model for the division of the task of urban rail transit is established according to the regulations of the manual operation of the urban rail. This paper proposes a CGLR algorithm based on the idea of column generation, and uses this idea to obtain the sub-set of small-scale optimal tasksets to reduce the complexity of solving the task partition problem. Taking the optimal Lagrange multiplier as inspiration The information LR_Heuristic algorithm replaces the simplex algorithm to solve the RTSCP relaxation problem and improves the efficiency of the algorithm. By combining the Lagrange multipliers obtained, this paper introduces a stochastic column repair technique to obtain a feasible solution to the RTSCP problem and improve the solution quality. Finally, the subway line to verify the background. The results show that the model and the algorithm can effectively solve the problem of crew assignment and get a better classification scheme.