A Novel Approach to Supplier Selection in a Supply Chain Management

[[abstract]]供應鏈管理的主要目標是將客戶的需求和供應商的原物料供應同步化，使其能在提供高品質的顧客服務、低廉的投入成本以及單為成本降低上取得平衡。快速回應訊息的目標透過網際網路可以達到在供應商和買家之間取得較佳的平衡。為了達到高顧客服務標準、低成本的目標，其中廠商選擇問題對於資訊的需求由以為甚。然而，廠商選擇通常包含了不同屬性的資訊需要分析，例如定性與定量的資料，傳統的統計方法對於處理有此特性之資料是難以達到預期目標的。其他幾項分析方法如資料包絡法、階層分析法亦不易對定性資料進行處理分析。 在本研究中，主要針對了多原物料與多供應商問題間提出混合理論用以解決廠商選擇問題並找出廠商營運趨勢以選擇較佳的合作廠商。這個混合式理論結合了約略集合理論與模糊時間序列分析，其主要目標在於預測廠商未來的生產與營運狀況，再分析成本用以選擇較佳的長期合作廠商。最後期望能選擇出能提供高生產效率以及低單位成本的廠商組合目標。[[abstract]]The goal of managing a supply chain is to synchronize requirements of the customers with the flow of materials from suppliers in order to strike a balance among conflicting goals of high quality of customer service, low inventory, and low unit cost. The aim of quick-response information can achieve the balance easily through exchanging desired information between suppliers and buyers through Internet. Specifically, the desired information to aim the supplier selection is crucial in providing high quality of service, low inventory, and unit cost. However, the criteria of selecting supplier may involve in processing qualitative information that is difficult to be analyzed by standard statistical techniques. The traditional methodologies such as Data Envelopment Analysis (DEA) and analytical hierarchical process (AHP) are difficult to handle the qualitative type of data processing. In this thesis, it is assumed that are multiple materials in manufacturing processes and suppliers are involved in this supply chain. The proposed hybrid model which incorporates rough set theory (RST) with the fuzzy time series (FTS) analysis aims to select desired suppliers for each manufacturing process such that the selection of aggregated optimal suppliers for high production performance and low production cost could be achieved.[[note]]碩