An Efficient Controlled Elitism Non-Dominated Sorting Genetic Algorithm for Multi-Objective Supplier Selection under Fuzziness-

Managers in supply chain management environments have to make important strategic decisions about supplier selection and order allocation. In a two-level supply chain having multi-period, multi-source, and multi-product characteristics, we propose in this work a multi-objective fuzzy model for supplier selection and order allocation. Along with coverage and weight optimization, the supplier evaluation goals underlined in this model include cost, delay, and electronic-waste (e-waste) minimization. The price discount given by the suppliers is modeled using a signal function. The uncertainty of delay and e-waste parameters is addressed using triangular fuzzy numbers; the weights of the suppliers are obtained using the fuzzy technique for Order Performance by Similarity to Ideal Solution (TOPSIS). Developed is the resulting NP-hard problem, a Pareto-based meta-heuristic algorithm known as controlled elitism non-dominated sorting genetic algorithm (CENSGA). The applicability of the CENSGA algorithm and the Taguchi technique is validated using the Non-dominated Sorting Genetic Algorithm (NSGA-II) and Multi-Objective Particle Swarm Optimization (MOPSO), so optimizing the parameters of the algorithms. Graphical and statistical comparisons showing how the proposed CENSGA dominates NSGA-II and MOPSO when it comes to of mean ideal solution distance (MID) and spacing metrics help to analyse the results.