| 摘要: | 良好的航機修護計劃可幫助航空公司有效率地完成航機修護工作，以確保航機飛行安全及航次營運順暢。然而，以往航空公司對於修護人員的排班，大多以人工經驗進行，不僅費時且往往無法配合實際營運之修護需求，造成人力的浪費。此等問題的相關文獻相當罕見，目前雖有初步的相關研究，然其規劃方式未考量目前實務上常採用之策略，以致效果有限。另外，近年來國內部分企業已逐漸採用國外盛行之彈性化的管理方式，如「工作時間彈性化」及「數量彈性化」等新觀念，以提高競爭力。有鑑於此，本研究針對航空公司停機線修護人力供給之問題，利用數學規劃及電腦演算法技巧，配合可行的彈性管理策略，構建合適之航機修護人力供給規劃模式及解法，幫助航空公司有效率地規劃修護人員之排班。 本研究首先依國內一航空公司現行之修護人員排班方式建構一基本模式，而後考慮不同排班彈性策略：彈性班次、彈性修護小班人數及彈性工作時間長度等，發展不同策略模式。此等模式可定式為混合整數規劃問題，屬NP-hard性質的問題。由於本研究所構建之各模式規模甚為龐大，在合理的時間內難以求得最佳解，因此本研究進一步發展新的啟發解法以幫助求解。在此啟發解法中，本研究利用C程式語言撰寫輸入檔轉換程式，並搭配使用CPLEX套裝數學規劃軟體協助設計。最後，本研究以該航空公司的實際修護資料為例，進行實例測試與分析。 A good maintenance plan can help an airline to efficiently perform aircraft maintenance so as to ensure its aviation safety and flight punctuality. However, in the past the airline maintenance scheduling was mostly performed by staff based on experience, which is not only time-consuming but also ineffective in meeting the maintenance requirements, resulting in waste of manpower. There was very little literature dealing with such problems in the past. Although a preliminary research was carried on the problem recently, it did not consider practical strategies, limiting its effect in practice. On the other hand, flexible management strategies from overseas, for example, flexible working hours and flexible manpower, have been recently popular and applied by some carriers in Taiwan. In this research, we use mathematical programming techniques and computer algorithms, incorporating feasible flexible management strategies, to develop suitable models and solution methods, in order to help airlines efficiently and effectively plan their maintenance schedules and manpower supplies. We first developed a basic model. Then incorporating different flexible strategies such flexible shifts, flexible crew members and flexible working length, we developed several strategic models based on the basic model. These models are formulated as mixed integer programs. Since their problem sizes are huge and cannot be optimally solved in a reasonably short time, we developed a new solution algorithm to solve the problems. We used C computer language to code all necessary programs and apply the mathematical programming solver, CPLEX, as well to develop the solution algorithm. Finally, to evaluate the models and solution algorithm, we performed a case study using the real operating data from a major Taiwan airline. |
