行李卸載系統為機場營運之重要設備，主要目的為準確地將行李送上飛機且不延誤各班機之起飛時間，使機場之營運順暢。隨著廉價航空的興起及紅眼班機的增加，某國內機場營運者基於日益增加的旅客量，為了因應龐大的運輸，於2015年將第二航廈境部分原先南北座的直線型行李卸載道全面更新成轉盤型行李卸載轉盤，以提升地勤人員作業之效益。在實務上，當行李卸載轉盤遭遇突發性故障時，未以系統最佳化之方法，仍以人工經驗進行調整其排班，不僅難以於短時間之內進行調整其航班規劃且更容易有缺乏整體性規劃及效率之情況產生，進而導致成本增加與機場資源浪費。因此，如何在短時間內有效地調整行李卸載轉盤之指派，使擾動狀態能盡快地恢復正常，使班機準確地依照預定班表飛行及航空公司的地勤人員是否正常作業，實乃當前重要之課題。 本研究利用數學規劃之方法，針對機場出境旅客行李卸載轉盤突發性故障之問題，以最小化所有指派航班之擾動時間加上航班作業重疊時間為目標，構建機場出境行李卸載轉盤突發性故障航班重新指派之模式。於求解方面，由於問題規模甚大，本研究利用撰寫基因演算法結合2-OPT區域搜尋法發展一啟發式解法以求解模式。為測試模式之實用性，本研究將以國內某國際機場之現有機場行李卸載轉盤設施及營運狀況數入資料，進行範例測試且針對不同參數進行敏感度分析，並進而探討實務上在設計最佳化模式及求解演算法時應注意之事項。本研究期望能幫助機場決策者在出境旅客行李卸載轉盤故障時能即時且有效地調整航班與行李卸載轉盤間之指派，以降低航班之延誤、及維持機場服務水準，並將此結果提供給學術界相關研究之參考。 ;The baggage unloading system is an important equipment for airport operations. Its main purpose is to accurately send baggage to the aircraft and not delay the departure time of each flight, making airport operations smoother. With the rise of low-cost airlines and the increase of red-eye flights, a domestic airport operator has based its ever-increasing number of passengers. Updated to a turntable-type baggage unloading dial to enhance the effectiveness of ground crew operations. In practice, when the baggage unloading carousel encounters a sudden failure, it is not the method of optimizing the system. It still adjusts its schedule with manual experience. It is not only difficult to adjust its flight plan within a short time, but also it is easier to have the lack of overall planning and efficiency results in increased costs and wasted airport resources. Therefore, how to effectively adjust the assignment of the baggage unloading carousel in a short period of time so that the disturbance status can be restored to normal as soon as possible, so that it is an important issue for the flight to accurately fly according to schedule schedules and whether the ground crew of the airline is operating normally. In this study, a method of mathematical planning is used to solve the problems about airport passenger baggage unloading carousel malfunctions abruptly that failures to minimize the disturbance time of all assigned flights and the overlapping time of flight operations. In terms of the solutions of problems, the proposed model is characterized as NP-hard. Moreover, In order to address the large scale problem efficiently, a genetic algorithm, coupled with a 2-OPT local search, is developed. Besides that, we test the practicability of the model by using a case study including practical information from an international airport in Taiwan to evaluate the model and the heuristic algorithm. This study is expected to provide airport decision makers a useful method to immediately and effectively adjust the assignment between flights and baggage unloading carousels when the outbound passenger baggage unloading carousel fails. This result is able to be used as a reference for relevant researches in the academia.
