RFID network planning for wireless manufacturing considering the detection uncertainty

The RFID reader-tag detection uncertainty comes from inherent unreliability of the commercially available hardware due to the collisions between RFID devices and interference from the surrounding environment. The performance of an RFID network would be largely impacted by the detection uncertainty which should be considered in the network planning phase to minimize its negative impact. In this paper, we establish a non-linear integer programming (NLIP) RFID network planning model for a mixed-model assembly line considering the detection uncertainty. The model minimizes the total cost by determining the optimal number of readers with their locations and types subject to the given detection reliability requirements. Then, a genetic algorithm (GA) integrating a divide-and-conquer greedy randomized adaptive search procedure (DCGRASP) is designed to solve the problem. We further apply the proposed model and algorithm to an industrial case of an automobile mixed-model assembly to illustrate how the model can meet the needs of the wireless manufacturing. The results of the case study also demonstrate the effectiveness of the proposed algorithm.