Hybrid adaptive simplified human learning optimization algorithms for supply chain network design problem with possibility of direct shipment

Abstract

A two-stage supply chain network design problem with a minimization type cost-based objective function is focused in this study. Some important assumptions such as considering different transportation modes in the stages, fixed charge transportation cost, and possibility of direct shipment between plants and customers are respected in the problem. As the problem is of NP-hard class of optimization problems, some meta-heuristic algorithms are proposed as its solution approach. In this regard, the recently proposed adaptive simplified human learning optimization (ASHLO) algorithm is used to be hybridized by the genetic algorithm (GA) and Particle swarm optimization algorithm (PSO) separately. In addition, two more meta-heuristic algorithms of gravitational search algorithm (GSA) and cuckoo search (CS) are used for more comparisons. Therefore, totally thirteen classical and hybrid meta-heuristic algorithms are proposed to solve the problem. In the extensive computational experiments of the study, 40 test problems from various sizes are generated. A typical experimental study is done to tune the parameters of the proposed algorithms. Using the results of parameter tuning step, the final experiments on the test problems are performed. The obtained results prove the superiority of the ASHLO algorithm when hybridized by the PSO.