良好的道路養護工作隊排程規劃，可以提升養護效率，並降低成本。目前國內養護廠商大多依照以往經驗，估算平均的旅行時間與修護時間，以進行工作隊排程規劃。此作法忽略了實務營運上，旅行時間與修護時間之隨機特性，若隨機性擾動過大時，則可能使原規劃的結果失去其優越性，亦即最佳化工作隊排程結果可能不為實際最佳排程。此等最佳化道路養護工作隊排程在營運中受到隨機因素擾動的影響，在過去未曾發現有文獻進行探討。有鑑於此，為規劃較符合實務情況的工作隊排程，本研究考量實際營運時，旅行時間與修護時間之隨機變動狀況及實務營運的相關限制，以養護廠商總未養護時間最小化為目標，構建一確定性與一隨機性工作隊排程模式，幫助廠商在有限的養護資源下作出最佳的工作隊調派決策，並提高養護作業效率。 本研究參考國內實際的養護工程作業方式，利用時空網路流動技巧，以系統最佳化的觀點，構建一確定性道路養護工作隊排程模式。之後，進ㄧ步修正確定性模式中固定之旅行時間與修護時間為隨機變數，構建一隨機性道路養護工作隊排程模式。求解上，本研究以CPLEX數學規劃軟體直接求解模式。另外，為評估隨機狀況下的作業績效，本研究並將發展一模擬評估方法，以比較實務排程、確定性與隨機性排程規劃的結果於實際營運應用之差異。最後，為測試本研究模式與評估方法的實用績效，本研究以某縣市政府所負責之養護範圍為例，利用C程式語言結合數學規劃軟體CPLEX進行實例測試與分析，進而提出結論與建議。 A good work team schedule for road maintenance can help improve the effectiveness of the road maintenance and thus reduce the operating costs. In current practice the work team schedule is typically designed by the staff’s experience, in accordance with the projected (or average) fleet travel times and maintenance times, meaning that stochastic disturbances arising from variations in vehicle travel times and maintenance times in actual operations are neglected. In the worst case scenario, where vehicle travel times and maintenance times fluctuate wildly during daily operations, the planned work team schedule could be disturbed enough to lose its optimality. Since there has been no research on work team scheduling problems that can account for stochastic fleet travel times and maintenance times, in this research stochastic disturbances of daily vehicle travel times and maintenance times that occur in actual operations are considered from the basis of the carrier’s perspective. We try to develop a deterministic and a stochastic work team scheduling models, with the objective of minimizing the total operating time (including total vehicle travel time and total maintenance time). The models are expected to be useful planning tools for carriers to decide on their optimal work team schedules in their operations. We will employ network flow techniques, with a systemic optimization perspective, to construct a deterministic work team scheduling model. Then a stochastic work team scheduling model is developed by modifying the fixed vehicle travel time and maintenance time parameters in the deterministic work team scheduling model. We use the mathematical programming solver, CPLEX, to directly solve the scheduling model. To evaluate the models and the solution algorithms under stochastic vehicle travel times and maintenance times in actual operations, we will also develop a simulation-based evaluation method. Then, we will perform a case study on personal computers, using real data of a local government. Finally, conclusions and suggestions will be given.
