부산대학교 도서관

The Kanban system driven by RFID technology for identifying individual products, has been developed to deliver the finished products to meet the customer's actual requirements and improve the automated production process at the manufacturer. However, the developed model focuses entirely on the IT infrastructures, thus placing a financial burden on small and medium size enterprises. To eliminate this problem, a cloud-based infrastructure for a Kanban-RFID system with low costs through reducing the waiting time to enter into the production process is proposed. The proposed Kanban-RFID model is divided mainly into 2 layers; the enterprise and cloud computing layers. In each layer, the primary functions and communications of the service components were determined so that the Kanban-RFID system could correspond directly to the mechanism of the production process and economically place it on both IT and cloud computing infrastructures. To evaluate the performance of the proposed system, rice mill manufacturing was used as a case study in order to find 2 criteria metrics. Firstly, the results evaluated through the simulation model using Arena simulation showed that the amount of waiting time spent on the operational process of the proposed model, compared with the traditional model (push system), decreased from 201.65 to 88.19 min (56.26%). This output subsequently increased the number of finished goods delivered to the customers from 615 to 651 (5.73%). Secondly, the cloud-based proposed system gained 18.27% of the break-even point at 1 year 7 months and 27.14% of the return on investment. [ABSTRACT FROM AUTHOR]