利用車輛自動監控系統資料評估貨車司機行徑績效之研究

Using AVL System Data to Evaluate Truck Drivers' Routing Performance

鑑於如何評估司機在公司外作業是否合理，一直是貨運業者所關心的問題；本研究利用車輛自動定位（AVL）系統資料，提出一個評估司機合理行駛路徑距離及合理停等地點、時間與次數等行徑績效指標之分析模式，進而建置汽車貨運業司機行徑績效管理系統。受限於AVL系統及資料庫中訂單收、卸貨地點座標之不足，本研究提出多項修正方法與判斷準則來提高司機行徑績效評估準確度；使汽車貨運業者可藉由該績效評估系統有效的輔助管理司機行駛路徑與建立明確的貨車司機服務績效獎懲管理規範。最後利用國內某貨運公司3個月份實際營運資料作為分析範例，其結果顯示公司行徑績效較差的前1／3的司機，由於行駛路徑或停等時間的不合理，平均每位司機造成公司在車輛油料、保養、維修等方面，每月額外支出約新台幣六千至一萬元，對貨運公司而言是一筆非常可觀的額外支出。由範例的驗證，本研究所建置之司機行徑績效管理系統確有助於提升貨運業者之競爭力，也提升業者採用車輛自動定位系統之意願。

Since it is very difficult to evaluate drivers' performance outside the company, we proposed several driver routing performance indices to help trucking company managers evaluate their drivers. In addition, a driver routing performance evaluation system was developed by using Automatic Vehicle Location (AVL) system data. In this system, we especially developed a model to judge whether the routing, stop locations, stop frequency, and stop durations of each trip for each driver are reasonable. Due to the limitation of data accuracy obtained from the Global Positioning System and the scarcity of location data for new customer orders, we proposed several methods to increase the evaluation accuracy of driver performance. Real world operational data for three months of a private trucking company in Taiwan was analyzed by the system. The results indicate that the loss of this company due to the poor performance of drivers in the worst 1/3 is NT$6 ~ NT$10 thousand dollars per driver per month. The proposed system can help the trucking company evaluate its drivers accurately as well as reduce its operating costs.