بهبود اثربخشی موجودی در زنجیره تامین جهانی آر اف آی دی فعال با مدل های پیش بینی خاکستری

A RFID-enabled global TFT–LCD supply chain associated with Grey forecasting model (GM) of Company A has been simulated and analyzed in this research. Three key performance indicates (KPI) including total inventory cost, inventory turnover and bullwhip effect are analyzed in the simulation experiments in order to compare the effectiveness of five different supply chain inventory models. The effectiveness of integrated system which is composed of supply chain operation, Grey short-term forecasting model and RFID system has been examined by aforementioned three KPIs. According to the result of Taguchi experiments, RFID-enabled R-SCIGM supply chain model which integrates the GM(1,1) forecasting model based on (s, Q) pull-based replenishment policy reduces 43.36% of the total inventory cost compared with that of the non-RFID SCIGM model. It apparently shows that a great improving effectiveness of supply chain inventory cost can be conducted while RFID system is incorporated with the GM(1,1) forecasting model.

There has been fierce competition between companies under the trend of globalization of industries. Therefore, the problems of reducing the length of product lifecycle, tightening shipment date and quick delivery of products are faced by the industries. In order to achieve the ultimate value of integrated operations, industries are going to focus on the integration of supply chain. The interrelated networks of supply chain consist of manufacturers, suppliers, distributors, wholesalers, retailers and customers. The volume of demand forecasted by upstream suppliers is usually greater than the actual market demand, for the lack of transparent information and communication channel. The wrong decisions on inventory and production are caused by the distorted information on purchasing products. Thus, the increased inventory cost in the upstream channels leads to the enlargement of the total cost of entire supply chain, which creates the bullwhip effect. Lee et al. (1997) considers demand forecasting, lead time, price variation and batch orders as factors of bullwhip effect. He thinks one of the ways to weaken bullwhip effect is to avoid repeated demand forecasting. To lessen the problem of bullwhip effect, the sharing and exchanging of supply chain information is needed. For achieving this goal, one way is to utilize the Radio Frequency Identification System (RFID). With the real-time product visibility and traceability of RFID, the amount of on-hand inventory across supply chain tiers can be precisely calculated, and the lead time of product delivery can be shortened. Moreover, the impact which results from errors in demand forecasting can be reduced, and the effectiveness of supply chain management can thus increase. In order to cope with the instant RFID-enabled information sharing capability, the Grey forecasting model can be appropriately adopted based on merely four pieces of short-term historical data. Therefore, there is a great opportunity for integrating the real-time data retrieval capability of RFID with short-term forecasting merit of GM to achieve the quick response in the operation of supply chain. The purpose of this research is to establish a simulation model of global supply chain, which integrates the Grey short-term forecasting model and specific service level as basic pull-based inventory replenishment policy for the TFT–LCD Company A, to prove that the implementation of RFID system can best improve the inventory cost effectiveness. The remainder of this paper is organized as follows: Section 2 provides a literature review on the RFID applications, supply chain simulation and Grey forecasting model. The global TFT–LCD RFID-enabled supply chain of Company A and simulation modeling is presented in Section 3. The Taguchi methods/design of experiments and verification by simulation are described in Section 4. In Section 5, the experiment results of simulation KPI output are analyzed and compared. The conclusion of this research is shown in Section 6.

From the result of the Taguchi experiments, we find that the optimal specification of R-SCIGM model is to implement the (s, Q) pull-based replenishment policy, GM(1,1) (α = 0.7) forecasting method associated with the RFID-enabled system. Based on the Bernoulli experiment, the R-SCIGM model has a 43.32% decrease in the total inventory cost in comparison to the SCIGM model. This shows that the adoption of RFID technology has a significant effect on the supply chain performance. The R-SCIGM has a 8.56% decrease in the total inventory cost in comparison to the R-SCIARIMA model. This also shows the adoption of the GM short-term forecasting method associated with RFID can effectively enhance the performance of forecasting function. The establishment of the R-SCIGM model in Company A can weaken the degree of bullwhip effect in the supply chain. In this research, we have successfully constructed a TFT–LCD industry-oriented simulation platform for evaluating the performance of the RFID-enabled global supply chain. The major mechanism for simulating the operations of global supply chain has been properly designed as a group of functional agents that can be sorted into ten categories. All of the input parameters of pull-based inventory replenishment policies are completely collected from Company A and then derived as decision variables using theoretical and practical equations. The GM-based customer demand forecasting model stemmed from the 52 weeks shipment data for TFT–LCD of Company A has been derived following six strict steps. We also have assessed the optimal factorial combination for pull-based replenishment policy, GM parameter and RFID utilization by adopting a well-known Taguchi experiment methodology. Finally, the experiment shows that the integration of RFID-enabled real-time information sharing process, pull-based replenishment policy (s, Q) and GM short-term customer demand forecasting model in the global supply chain can effectively enhance the effectiveness of its inventory cost management. For industries RFID promoters, our analytical evidence presented in this research can help them identify the implementation principle: RFID adoption must be cooperated with an appropriate inventory replenishment policy and demand forecasting approach.