為解決城市中的交通壅塞問題並降低對環境的傷害，政府極力倡導民眾使用大眾運輸工具，各都市亦積極規劃與發展大眾捷運系統，而隨著捷運路網不斷的興建與擴張，該如何維護系統品質並提供安全的乘車環境顯得至關重要。目前實務上針對捷運場站發生故障時的即時搶修人力調派及後續維修排程調整，主要仍仰賴人工經驗進行規劃，此種缺乏系統性分析的方式往往無法做出最即時且最低成本的決策。有鑑於此，本研究以捷運公司之角度，考量即時搶修人力調派及剩餘維修排程擾動，以總維修成本最小化為目標，構建一捷運場站維修排程即時調整模式，期能協助決策者即時且有效的調整維修排程及調派人力。 本研究利用時空網路流動技巧構建此模式，並透過C++程式語言結合數學規劃軟體CPLEX求解模式，另外運用滾動分析概念搭配動態決策以進行維修排程調整規劃及人力調派。最後為評估模式之實用性，本研究以國內某捷運公司之實際營運資料進行範例測試，並調整不同參數和規模進行敏感度與方案分析。結果顯示本模式能夠有效運用於實務上，可供捷運公司決策單位參考，以在未來進行維修排程調整規劃及人力調派。 ;In order to solve the traffic congestion problem in cities and reduce the environmental impact, the government has been promoting the use of public transportation, and cities are actively planning and developing the mass rapid transit (MRT) systems. With the continuous construction and expansion of the MRT network, it’s crucial to retain the system quality and provide a safe riding environment. In practice, the deployment of the crew for emergency repair and subsequent maintenance schedule real-time adjustment when facing system failure still relies on traditional manual experienced scheduling at present. This policy lacks systematic analysis and often fails to make the most immediate and least cost decisions. As a consequence, from the perspective of the MRT corporation, this study considers the crew assignment for emergency repair and perturbation of the subsequent maintenance rescheduling, and develops a real-time adjustment model for the maintenance schedule of the MRT stations with the goal of minimizing the total maintenance cost. It is expected to assist decision-makers in effectively planning maintenance schedule and crew assignment. This study utilizes the time-space network technique to build this model which is solved through the C++ computer language and the mathematical programming software CPLEX. In addition, this study employs the concept of rolling analysis with dynamic decision-making for maintenance adjustment and crew assignment. Finally, to evaluate the practicability of the model, a case study was conducted using the operational data of a domestic MRT corporation, and different parameters and scales are adjusted for sensitivity and scenario analyses. The results show that this model can be effectively used in practice and thus can be used as a reference for decision-makers of MRT corporations.
