A Self-Learning Particle Swarm Optimization for Robust Multi-Echelon Capacitated Location-Allocation-Inventory Problem

This paper addresses a multi-echelon capacitated location-allocation-inventory problem under uncertainty by providing a robust mixed integer linear programming (MILP) model considering production plants at level one, central warehouses at level two, and the retailers at level three in order to design an optimal supply chain network. In this model, the retailer's demand parameter is uncertain and just its upper and lower bounds within an interval are known. In order to deal with this uncertainty, a robust optimization approach is used. Then, a self-learning particle swarm optimization (SLPSO) algorithm is developed to solve the problem. The results show that the proposed algorithm outperforms the exact method by providing high quality solutions in the reasonable amount of computational runtime.