Wang Jiaxi, Gronalt Manfred, Sun Yan
School of Traffic and Transportation, Beijing Jiaotong University, Beijing, People's Republic of China.
Institute of Production and Logistics, University of Natural Resources and Life Sciences, Vienna (BOKU), Vienna, Republic of Austria.
PLoS One. 2017 Jul 13;12(7):e0181165. doi: 10.1371/journal.pone.0181165. eCollection 2017.
Due to its environmentally sustainable and energy-saving characteristics, railway transportation nowadays plays a fundamental role in delivering passengers and goods. Emerged in the area of transportation planning, the crew (workforce) sizing problem and the crew scheduling problem have been attached great importance by the railway industry and the scientific community. In this paper, we aim to solve the two problems by proposing a novel two-stage optimization approach in the context of the electric multiple units (EMU) depot shunting driver assignment problem. Given a predefined depot shunting schedule, the first stage of the approach focuses on determining an optimal size of shunting drivers. While the second stage is formulated as a bi-objective optimization model, in which we comprehensively consider the objectives of minimizing the total walking distance and maximizing the workload balance. Then we combine the normalized normal constraint method with a modified Pareto filter algorithm to obtain Pareto solutions for the bi-objective optimization problem. Furthermore, we conduct a series of numerical experiments to demonstrate the proposed approach. Based on the computational results, the regression analysis yield a driver size predictor and the sensitivity analysis give some interesting insights that are useful for decision makers.
由于其环境可持续和节能的特点，铁路运输如今在运送乘客和货物方面发挥着基础性作用。在运输规划领域出现的乘务员（劳动力）规模确定问题和乘务员排班问题，受到了铁路行业和科学界的高度重视。在本文中，我们旨在通过在电动动车组（EMU）车辆段调车司机分配问题的背景下提出一种新颖的两阶段优化方法来解决这两个问题。给定一个预定义的车辆段调车计划，该方法的第一阶段专注于确定调车司机的最优规模。而第二阶段被制定为一个双目标优化模型，其中我们综合考虑最小化总步行距离和最大化工作量平衡这两个目标。然后我们将归一化正态约束方法与改进的帕累托过滤算法相结合，以获得双目标优化问题的帕累托解。此外，我们进行了一系列数值实验来验证所提出的方法。基于计算结果，回归分析得出了一个司机规模预测器，敏感性分析给出了一些对决策者有用的有趣见解。
