隨著全球經濟與貿易蓬勃發展，旅遊觀光風氣鼎盛，航空運量大幅增加，使航空運輸成為不可或缺的交通工具之一，在偌大的航運量中飛航安全更為不可輕忽之課題，其中飛航管制對於飛航安全擔任極其重要之角色。而飛航管制之運作為24小時不間歇，且飛航管制人員執勤時不僅需長時間全神貫注，工作時亦承受相較於一般工作較大之心理壓力。然而，目前飛航管制人員排班規劃主要是透過人工方式進行，較缺乏整體系統性的分析與考量。有鑑於此，本研究以系統最佳化之觀點，航管單位決策者之立場，考量飛航管制人員執勤產生之疲勞感，發展一飛航管制人員人力排班模式，期望能輔助決策單位有效地規劃飛航管制人員人力排班。 本研究模式係利用數學規劃方法及時空網路流動方式，構建飛航管制員人力排班之時空網路，並定式飛航管制人員於時空中之排班，在滿足所有班別之需求人數和實務方面限制為前提下，以最平均化每位飛航管制人員工作總時數為目標。在求解方法上，利用C++程式語言配合CPLEX數學規劃軟體進行問題之求解。而後為評估模式之實用性，本研究以國內某航管單位為例進行測試，並針對不同參數進行敏感度分析，結果顯示本研究提出之模式在實務上能有效地運用，可提供決策單位作為飛航管制人員人力排班規劃之參考。 ;Air traffic controllers (ATCOs),who are essential to the aviation safety. ATCOs operate around the clock and work in an environment requiring high focus levels over long working hours, and are faced with higher psychological stress levels compared to those of most other professionals. Nowadays, domestic ATCOs shift scheduling is mainly arranged in a manual way, which lacks the perspective of system optimization and makes formulated ATCOs scheduling inefficient. Therefore, the aim of this study is to construct a ATCOs shift scheduling model where the actual air traffic controller scheduling constraints and ATCOs fatigue generated while on duty are taken into consideration. With this model, the decision maker can plan ATCOs shift schedules more effectively. In this study, a scheduling model of the ATCOs is developed by utilizing mathematical programming method and the time-space network flow technique. All demands of people in each work shifts must be satisfied, the related operating constraints are ensured. The model was aimed to average working hours of all controllers and expected to be an effective planning tool to assist the ATCOs in scheduling. The C++ computer language, coupled with the CPLEX mathematical programming software, is employed to solve the problem. The numerical tests are performed using the work shifts data associated with Air Traffic Control (ATC) unit in Taiwan. Computational results of case study were given for evaluating the performance of the proposed model. Finally, conclusions and suggestions made based on the computational results were given.
