Multiobjective Estimation of Distribution Algorithm Combined with PSO for RFID Network Optimization

A new multi-objective estimation of distribution algorithm combined with PSO by using a Pareto-based method is proposed and applied in RFID network design. In the algorithm´s offspring generation scheme, one part of individuals is sampled in the search space from the constructed probabilistic distribution model and the other part individuals are generated by the velocity-free PSO. A balance parameter is used to control contributions of the global information from EDA individuals and the local information from velocity-free PSO individuals. The nondominated sorting and ranking selection procedure are applied to constitute a nondominated solution set from the current population. The proposed algorithm is applied to some benchmark functions and a RFID network design problem. The relative experimental results show that the algorithm has better performance than some algorithms and is effective in the design of RFID networks.