隨著新的價值觀與消費型態的萌生，企業為降低中間轉運費用，節省管理及倉儲成本，使得物流中心的設立與需求，將成為儼然的發展趨勢。物流中心的主要功能係將商品集中並分散至零售業，目的在於增快貨品流通速度並且避免不必要的運儲成本，使商品能快速運送到達顧客手中。許多統計資料顯示，整個物流作業在公司的營運成本中，佔了相當大的比例。大多數的物料處理作業，都是屬於勞力密集且高重複性，而在倉儲作業中，產品的存放擺設位置以及揀貨作業策略，又直接地影響了物料處理成本的高低。 在物流中心的各項內部作業中，揀貨作業是一項重要且繁雜的工作，是否能在合理的時間內完成此項作業，直接關係著物流中心的經營成本與服務水準。物流成本大約佔商品最終售價的30%，但揀貨成本卻是其他堆疊、裝卸、運輸等成本總和的九倍，所以針對不同型式的訂單，應選擇一個適合的揀貨策略以提升整個揀貨作業績效 。因此，如何規劃一個合理的揀貨作業，有效的降低揀貨作業時間與成本，提昇貨品揀取與處理的作業效率，對於物流中心的營運績效將有決定性的影響。 本研究將針對「訂單批量揀取」問題做有系統的整理與比較；在兩種迥異走道被拜訪機率分配下產生兩種不同的揀貨清單，搭配上兩種大小不同的揀貨車容積，以八種種子訂單選取法則、八種配合訂單選取法則，以及兩種路徑規劃策略交互搭配實驗。吾人利用Visual Basic 6.0做模擬實驗，進行不同策略組合之下的實驗分析，對系統模擬產生出的實驗數據以Excel做輸出整理，並輸入統計套裝軟體SPSS10.0作一比較分析。評估揀貨旅程次數的多寡，以及比較揀貨路徑的長短，以期得到較佳的訂單批次組合方法，並對發展出之演算法作一結論，以及對未來研究方法的建議。以供實體物流中心的揀貨參考依據。 We look into one of the problems that one often encounters in the order picking of distribution centers. This problem is known as order-batching problem. In an order-batching problem, one has to select two or more orders, batch them, and pick their requested items at the same order-picking trip. Order-batching problem is an important problem in distribution centers that often have to handle many different orders in one day. This is because one can reduce the total travel distance if orders with similar picking locations are batched and picked together. And, the order-picking cost can be reduced as a result. The proposed order-batching method has two major steps to form an order batch. At the first step, a seed order will be selected from the order pool. A seed order is the first order added to an order-batch. We refer to the problem at the first step as “seed-order selection problem”. Once a seed-order has been selected and added to an order-batch, at the second step additional orders will be selected and added to the order-batch. We refer to the problem at the second step as “accompanying order selection problem”. We will propose different solutions for each problem. For the seed-order selection problem, we propose eight different seed-order selections rules. We also propose eight different selection rules for the accompanying-order selection problem. Furthermore, we also investigate two route-planning methods to arrange the order-picking route for an order batch. Test problems are then generated and solved by each one of them. Their solution results are collected and compared to see which seed-order selection rule, which accompanying-order selection rule, which route-planning method, and what combination of them can result in the best performance in minimizing the total travel distance. It is hoped that the knowledge learned from this study can provide substantial benefits to practitioners in distribution centers.
