Selective Regeneration Particle Swarm Optimization for inventory classification in supply chain systems

This paper presents Selective Regeneration Particle Swarm Optimization (SRPSO) for inventory classification problems in a two-stage supply chain where inventory items are classified based on a specific objective, minimizing total related costs. In addition, the proposed classification technique automatically determines the best number of inventory classes. This study employs an experiment design to determine the appropriate combination of algorithm parameter values, and tests a real dataset along with three article databases to compare the results to other known classification and non-classification methods. The outcomes fully demonstrate that SRPSO is an efficient, accurate, and robust method for inventory classification in supply chain problems. The SRPSO performs comparatively better than other grouping and non-grouping techniques.