| 摘要: | 近年來環保及資源回收再利用觀念逐漸受到注重，廢棄車輛回收產業也逐漸興起。然而，廢棄車輛車源皆為非固定時間、非固定位址及非固定產生量，故對於廢棄車輛回收業如何選擇最佳位址設廠以進行回收廢車服務，面臨極大問題。且油價近年來不斷上漲，對於回收業者而言，其運輸成本費用已占總營運成本近 18%，無法以傳統觀念之土地租金成本為選址主要考量因素（占總營運成本近 9%），故期望能發展一廢棄車輛回收廠區位選址模式，可望協助廢棄車輛回收業者能降低營運成本外，更能提昇回收廠服務績效與品質。 另外，本研究遂加入路段成本權重觀念，主要因過去有關運輸距離計算方式，大多僅考量所求解最短路徑之距離，但其結果於實際路網上往往面臨車多擁擠現象，進而導致運輸成本並非絕對最小化。因此，藉由此觀念實際應用於真實路網上，利用地理資訊系統軟體結合最短路徑之A* 演算法進行求解，再透過時空網路流動技巧規劃各候選廢棄車輛回收廠車輛排程作業，並以C++電腦語言撰寫演算法配合CPLEX數學規劃軟體求解。為了驗證所發展模式在實務求解上可行性，本研究將以桃園縣中壢市為例進行規劃求解，規劃服務需求點之最佳廢棄車輛回收廠位址。 In recent years, the idea of environmental protection and resources recycling and reuse have gradually concerned about, the industry of the abandoned vehicle recycling is also risen. However, it is unstable for the occurrence of abandoned vehicles in numerous ways, like time, location, and amount. Therefore, it’s a big problem when selecting the best site for building recycling plant to recycle abandoned vehicle. Since the oil prices keep going up in recent years, the transportation costs have been about eighteen percent of total operation costs for the recyclers. It can’t be the main concern that choosing the address by the traditional concept of renting land costs (occupying about nine percent of total operation costs). Then, we developed a site selection model which is deserted vehicles recycling center. The framework is expected to be useful for reducing the recyclers’ operation costs, and promoting the service performance and quality in the recycling center. In addition, we accede to an idea about the weight of section road cost, mainly because the method which relevant to calculate the transportation distance was used to solve the short path distances problems. However, the consequences confront the problem of crowded vehicles in realistic network, and it bring about the cost of transportation which is really not absolutely minimum. Hence, we used geographic information system with A* algorithm of the shortest path, as well as employing network flow techniques to construct a scheduling model which choose the recycling center. To efficiently solve the problem, we developed a heuristic, used the C++ computer language to write the necessary programs, coupled with the CPLEX mathematical programming solver. Then, to evaluate feasibility of the model and solution of realistically problem, we perform a case study by using real data of Jongli City, Taoyuan County to site the best location of plant. |
