摘 要 在傳統營造業，剩餘土石方清運車輛派遣上，多憑經驗法則做適當之調派，當案件量多且車輛需求排列複雜時，決策者的調派決定適當與否就成為重要之課題，若調度佳可提升效率而且利潤更好。本研究係探討所承攬的數個清運工程，以達最小成本的車輛派遣為目標，求出各案車輛資源指派，提供決策者調度運籌之參考；此外，依據各參數之敏感度變化趨勢，進行敏感度分析，藉此求得所有案件指派最小成本，達成效率及車輛人力機具充分運用之目的。求解方法上，本研究將各案件需要清運數量、可收容數量、工時限制及運輸時間等資料，利用EXCEL 軟體列出各案件的已知資料等基本資料，輸入LINGO 8.0套裝軟體求解，即可得最小成本及運輸指派總趟次等結果。 為驗證本模式合理性與實用性，本研究以自訂假設數據為範例，進行模擬測試，並以最佳化模式所求得之指派結果與一般實務經驗指派結果進行比較，結果發現確實較人工經驗指派來的快，且更有效率。顯示本研究可靈活運用於人員車輛先期排程規劃問題，幫助決策者有效地規劃剩餘土石方之清運作業。Abstract The deployment of truck fleet in the moving of residual earth relies mainly on the rule of thumb to decide appropriate vehicle dispatching in the traditional construction industry. If the caseload of contractor increases and the vehicle dispatching is more complex, whether the vehicle dispatching by the decision-makers is appropriate or not become an important issue for contractors. Good deployment can improve efficiency and make more profits. This study proposes a linear optimization model which determines individual vehicle resources required to be assigned by minimizing cost when several earth-removing works need to be processed by the contractor at the same time. This linear optimization model can be a reference for decision-makers. Besides, based on reviewing the variation trends of sensitivity in each parameter, the sensitivity analysis is able to find a solution with a minimum cost. This optimum allocation can not only improve efficiency but also make full use of vehicles as well as human resource. There are several basic data need to be tabulated by EXCEL software in the linear optimization process, such as the earth removing quantity for each work, the allowable quantity of dumping soil, working hour limitation and time of transportation. Using the input table from previous step as known data and solved with LINGO 8.0 software package, both a vehicle dispatching solution with minimum cost and the overall required trips for earth-removing can be determined. To verify the rationality and the usability of this mode, a set of customized data were assumed as basic data for several earth-removing works during simulation. The result obtained from this linear optimization model is compared with that obtained from method based on human experience. It is shown that this linear optimization model is more efficient and requires less processing time in comparison with manual allocation method which based mainly on human experience. As a result, the proposed linear optimization model can be flexibly applied to the pre-planning allocation of both human and vehicles resource. Furthermore, vehicle dispatching in the moving of residual earth cab be more effectively by planning with this proposed method.
