- 學位論文
輸送帶輔助揀貨系統之建議作業指派推論模式
A Task Assignment Model for Conveyor-aided Picking System
摘要
由於自動化技術及資訊科技之發展與普及，過去以人工作業為主之傳統物流產業亦隨科技與技術之發展，開始導入自動化（如分類機）或半自動化（如輸送帶系統）之輔助揀貨設備，以期能節省龐大的人工成本支出，進而提高物流中心競爭力。然對於利用輸送帶作為輔助揀貨工具之半自動化揀貨系統而言，如何決定輸送帶中各工作站所應負責之揀貨品項，以達到輸送帶中各工作站之作業負荷平衡，進而提升物流中心整體資源利用率，乃為揀貨作業規劃人員所關注之議題。傳統物流中心針對此類議題往往僅以經驗法則或決策者主觀認定方式為之，但由於人工決策過程耗時且主觀認定之決策結果未必為正確結果，因此可能造成揀貨效率不佳，甚至必須重新進行揀貨作業指派，而徒增整體物流作業成本。故本研究即針對輸送帶輔助揀貨系統之作業指派問題，提出一套「揀貨作業平衡方法論」，以透過電腦系統快速產生一符合專家期望之最適指派方案建議，作為揀貨規劃人員進行作業指派之參考。 於揀貨作業平衡方法論中，本研究乃先以揀貨日程總表中之待揀貨品品項為基礎，利用窮舉法或隨機方式產生候選作業指派方案。之後，透過與資深揀貨作業規劃人員訪談，以釐清其進行揀貨作業指派時所使用之經驗法則與精神意涵；其後，再行將經驗法則量化為經驗指標；並利用歷史作業指派資料，推論符合專家期望之經驗指標權重，以作為後續評選作業之參考。此後，透過前述步驟所得之經驗指標與其對應權重，由眾多候選方案中評選一最適指派方案，以作為揀貨作業規劃人員進行作業指派之參考。本研究除發展揀貨作業平衡模式與方法論外，乃依此方法建構一套「輸送帶作業指派支援系統」並進行案例驗證，進而確認方法論與技術之可行性。 整體而言，本研究所提出之揀貨作業平衡方法論將可有效降低揀貨作業規劃人員於作業指派所耗費之時間，並提升其決策結果之正確性，進而降低物流中心整體營運成本，達到提升物流中心競爭力之目的。
並列摘要
In the conveyor-aided picking system, the shop-floor supervisor has to assign the picking tasks to workstations on conveyor based on loading balance between the workstations. In the past, the traditional distribution center that utilizes the conveyor-aided picking system rely on expertise of the supervisor to balance the tasks of workstations. The traditional decision process for workload balance is usually time consuming and inefficiency. In order to assist the supervisor to efficiently and accurately search a workload balance solution for workstations of the conveyer-aided picking system. This research proposed a quantitative methodology for workload balance by extracting the expertise of domain experts. Based on the proposed methodology, a computer-aided decision support system can be developed to automatically generate the workload balance solutions of the conveyer-aided picking system. The main idea of this research is to extract the empirical principles by interview with the domain experts and to establish quantitative indicators for workload balance. In addition, weightings of indicators can be derived from historical information. By utilizing the indicators for workload balance, the supervisor can objectively and efficiently determine the picking tasks of workstations. As a whole, this research provides a quantitative methodology and technique for the supervisor to efficiently and accurately determine the picking tasks of workstations in the conveyor aided picking system.