即時交通資訊下車輛配送模擬評估架構之建立

An Evaluation Framework for Dynamic Vehicle Routing Problems with Real-time Traffic Information

物流管理中，商用車輛的路線與排程規劃為物流配送中心重要營運決策，一般將之歸納為車輛路線問題（VRP）之研究。在實際的運輸系統環境下，道路的行駛速率會隨著交通狀況變動而改變，道路旅行成本並非為固定值，因此如何在即時交通資訊下重新規劃車輛路線，降低車輛的旅行成本便成為一個重要的課題。本研究以掃瞄法為基礎構建車輛指派，採用禁制搜尋法更新車輛路線，並構建動態模擬評估架構。模擬評估架構中結合車輛指派演算法、路線更新演算法與動態交通模擬指派模式（Dyna TAIWAN），透過Dyna TAIWAN產生即時旅行時間矩陣，進行即時資訊下動態車輛路線問題之評估與分析。本研究利用50節點路網與臺中路網進行數值實驗，說明模擬評估架構的特性與可能的效益。

Logistics management has been one of the most important elements in supply chain management. Major daily operation issues in distribution centers are routing and scheduling of trucks. The problem to optimize vehicle routes is called the vehicle routing problem (VRP). The dynamic vehicle routing problem (DVRP) is an extension of VRP aiming to consider possible variations of travel times in the network. In this research, a two-stage framework for solving the dynamic vehicle routing problem is proposed. In this framework, the dynamic vehicle routing problem is divided into two stages, namely, vehicle assignment and vehicle routing improvement. In the first stage, a well-known approach, the sweep method, is conducted in vehicle assignment. In the second stage, a tabu search algorithm is implemented to improve route searching with real-time information. A heuristic approach is then applied in an evaluation framework in which assigning and routing operations could be simulated in a realistic traffic environment. The simulation-assignment model, Dyna TAIWAN is applied to evaluate assigning and routing strategies in a traffic network. Numerical experiments are conducted to illustrate the framework and possible benefits.