Optimal RFID Deployment in a Multiple-Stage Production System Under Inventory Inaccuracy and Robust Control Policy

Abstract

Deploying radio frequency identification (RFID) devices or applying robust production control policy can both reduce the negative impact of inventory inaccuracy. In this paper, we investigate the problem of optimally deploying RFID devices in a multistage production system with inventory inaccuracy and robust control policy. We first model the multiple-stage production system as a Markov decision process model, in which the decision of deploying RFID devices is included. After that, an integer programming model is constructed for optimally deploying RFID device problem. Meanwhile, we propose a robust production control strategy by employing the policy iteration and the decomposition/coordination techniques, and a heuristic algorithm is proposed to solve the integer programming model. Finally, numerical experiments are conducted to examine the performance of the proposed optimal RFID deployment approach. We discover that when RFID is deployed at a certain workstation (i.e., production equipment), the benefits of deploying RFID devices can transfer to the upstream and downstream stages (i.e., reduce the production cost of both upstream and downstream stages). Furthermore, numerical experiments reveal that the deploying RFID devices evenly among all the stages of the production system will generate a better performance.