| RFID(Radio Frequency Identification, RFID) technology has a great impact on the development of wireless technology and informatization.in recent years, it has always been one of the hot topics in the field of logistics research.Among all the problems of RFID technology, RFID network optimization planning is a very common but important one, whose purpose is to make its layout reasonable so that all the information stored in electronic labels in the work area can be read and wrote by readers in a concise manner. With the development of the concept of sustainable economic, closed- loop logistics network gradually become one of the core requirement of circular economy. there is a great difficulty in the realization of the closed- loop logistics, namely construction of a highly informative reverse logistics network, while the RFID technology has a great advantage in information capacity of tags and low dependence on working environment,the difficulties in the process of recycling of waste products,such as the complex and various source, poor working conditions and other problems, can be solved efficiently with RFID technology.it has far-reaching significance to the development of green industry chain to optimize the RFID network and apply it to reverse logistics.In previous studies, researchers have been using the evolutionary computation and population intelligence to solve the above problems, however, no matter the number of readers in the RFID system or the number of collection sites in reverse logistics network is difficult to predict. The proposed algorithm cannot give a suitable number, but the number of readers and collection sites make a difference in the complexity and operating costs of the whole system. Based on the traditional evolutionary algorithm in this paper, we have proposed the related improvement work, and this paper has done the main work as follows:1. For difficulties in predicting the number of readers in RFID network planning, this paper presents a multi-agent evolutionary algorithm with TRE operator to optimize RFID network planning.TRE operator tentatively deletes readers during the search process of MAEA and is able to recover the deleted readers after a few generations if the deletion lower tags coverage.by using TRE, the proposed algorithm is capable of adaptively adjusting the number of readers and moreover the coverage of tags is improved,and TRE also saves the computational cost. Finally through the analysis of the experimental results it can be seen that whether the coverage effect, or the number of readers, the proposed algorithm is superior to the traditional evolutionary algorithm, at the same time, the interference ratio of the RFID system is also optimized.2. Based on the in-depth research of proposed reverse logistics network, RFID technology has been applied into it.In this paper we propose a multi-swarm PSO algorithm to solve the problem of reverse logistics network planning. The algorithm contains a multi-swarm fusion factor, which can gradually reduce the size of swarm corresponding to an unreasonable number of collection points. When there is a lot of difference in performance in coverage between the optimum swarm and sub-optimal swarms, the size of these sub-swarms continue to decrease until it reaches zero. We also added RFID technology which can collect site information and transmit the statistical data to the data center. The simulation results show that the proposed algorithm can generate results with a high coverage rate and alternative number of collection points. |
PDF Full Text Request