整合RFID技術與雲端運算之智慧型產品銷售服務平台

Abstract

熱門商品在貨架上的流動性，將對門市的銷售業績產生關鍵性影響。對於零售業者來說，季節性或新聞性所引發的商品需求，只會維持一段時間的群聚效應與市場商機，倘若產品在上架銷售期間出現了供貨短缺，消費者將立即轉向其它商家，店家將因而錯失了商品銷售機會。 零售商店在商品需求規劃上，主要是透過銷售點系統(POS)來管理商品的銷售狀態，並利用供應鏈管理來改善長鞭效應的存貨資訊瓶頸；然而，熱門商品更具有時間效用，傳統系統架構與補貨模式所反映的庫存狀態，並無法呈現商品在賣場中的實際上架數量。因此，為防止資訊落差所造成之損失，我們提出一套「智慧型服務整合平台」，以軟體代理人做為基礎框架來建構整合性的資訊系統機制，並利用無線射頻辨識技術(RFID)設計智慧型商品貨架以做為商品訊息擷取之觸發始點，同時透過SOA機制部署跨平台的軟體元件，開發互動式資訊平台以提供商品的諮詢及推薦服務。 本論文的主要貢獻在於提出一個植基於雲端運算服務平台的方法論，結合RFID技術做為雲端前台架構(IT for Cloud Client)以及軟體代理人與SOA機制做為雲端後台服務(IT for Cloud Service)。為驗證所提出方法論在實務應用上的可行性，本研究利用所開發之雛型系統來進行效益評估，實驗結果也證明『整合RFID技術與雲端運算之智慧型產品銷售服務平台』整體架構能夠增進門市營運效能與提昇顧客服務品質，同時解決熱門商品的時間效用問題。

The mobility of the hot-sale product on the shelf will have a critical impact for the outlet sales. For retailers, the product demand triggered by the season or news will last only a period of time for clustering effect and market opportunities. If the product supply shortage occurs during the sales period, consumers will immediately turn to other vendors, so the store will thus miss out the sales opportunity. The Retail store, in view of the product demand planning, manages the product sales mainly through the point of sale system (POS) and uses the supply chain management to improve the bottleneck of the inventory information due to the bullwhip effect; however, the hot-sale product has more critical time utilization and the inventory status reflected by the traditional system structure and replenishment model cannot show the actual quantity of products on the store shelves. Therefore, in order to prevent losses caused by the information gap, we propose the “intelligent service integration platform”, in which the software agent serves as the basic framework to construct an integrated information system mechanism, the radio frequency identification (RFID) technology designs the intelligent product shelf as the trigger point for extracting the product message, and the SOA(Service-Oriented Architecture) mechanism is used at the same time to deploy the cross-platform software components to develop an interactive information platform so as to provide the product consulting and recommendation service. The main contribution of this paper is to propose a methodology based on the cloud computing service platform, which is combined with the RFID technology as a frontend cloud architecture (IT for Cloud Client), as well as the software agents and SOA mechanism as a backend cloud services (IT for Cloud Service). In order to verify the feasibility of the proposed methodology applied in the practical application, this study uses the prototype system developed to proceed with the benefit evaluation, and the experimental result also proves that the overall architecture of “RFID with cloud computing for intelligent sales service platform" can enhance the outlet operating performance and improve the customer service quality more effectively, while addressing the time utilization problem of the hot-sale products.