觀光Scan-Buy-Deliver模式之集貨與配送最佳化

Abstract

我國名特產商品，如太陽餅、鳳梨酥等，對觀光客極具吸引力但不便利之購買因素會影響觀光客購買意願，而Scan-Buy-Deliver (SBD)商店或稱虛擬商店，可降低觀光客購物之不便利性，且為我國中小企業提升營收之重要課題。 觀光客透過SBD購物模式購買名特產商品流程，觀光客於實體店面、飯店等，透過智慧型裝置掃描名特產商品之QR code，並於網頁瀏覽選購商品後，名特產業者接單後委託第三方物流公司，負責取貨與配送之服務，於位於郊區之名特產店倉庫、觀光工廠等地取貨後，進入城市中之虛擬集結點理貨後，配送至觀光客下榻之飯店、機場或交由旅行社。 因於城市中找到可供使用之廠站不易且耗費龐大成本，故本研究提出虛擬集結點，為城市中之臨時卸貨區，透過虛擬集結點之使用，取代屬於中長期之廠站選址問題，快速反應觀光客之需求，並降低成本。 本研究基於此情境，針對第三方物流公司，考慮多商品及時窗，並依取貨點是否供應重複商品或是否允許重複拜訪，區分出三種情境，建構取貨與配送最佳化之選址路徑問題數學模型，目標為最小化成本。 本研究發展基因演算法及模擬退火法，透過C語言撰寫程式，求解此NP-hard問題。再根據三種情境共設計10個範例，透過兩種演算法分別求解，比較求解能力之優劣。

Taiwan’s specialty products are extremely attractive to international visitors. Reducing inconvenient carrying, providing convenient and full services and reduction of logistics cost are the important factors for stimulating increasing consumer’s purchase intention and the small- and medium-sized retailers’ revenue. In the case of innovative SBD (Scan-Buy-Deliver) service, this study proposes an innovative operation model for freight consolidation and distribution. Through the proposed innovative operation model with virtual consolidation locations, it is anticipated that the logistics cost can be reduced and the logistics efficiency can be improved. This study attempts to construct a mathematical model to plan the freight consolidation and distribution. Taking the characteristics of cross-docking into consideration, the virtual consolidation locations and routes are optimally determined to minimize the total cost. Because the developed model is an NP-hard optimization problem, a Genetic Algorithm-based and a Simulated Annealing-based optimization approach are developed to resolve the freight consolidation and distribution model.Ten examples in different scales and situations are designed to solve and analysis.