Data driven hybrid evolutionary analytical approach for multi objective location allocation decisions: Automotive green supply chain empirical evidence

Abstract

The strategic location of manufacturing plants and warehouses and the allocation of resources to the various stages of a supply chain using big data is of paramount importance in the era of internet of things. A multi-objective mathematical model is formulated in this paper to solve a location-allocation problem in a multi-echelon supply chain network to optimize three objectives simultaneously such as minimization of total supply chain cost (TSCC), maximization of fill rate and minimization of CO2 emissions. Data driven hybrid evolutionary analytical approach is proposed by integrating Non-Dominated Sorting Genetic Algorithm-II (NSGA-II) to handle multiple objectives into Differential Evolution (DE) algorithm. Five variants of the hybrid algorithm are evaluated in addition to comparing the performance with the existing Multi-Objective Hybrid Particle Swarm Optimization (MOHPSO) algorithm. Extensive computational experiments confirm the superiority of the proposed Data driven hybrid evolutionary analytical approach over the existing MOHPSO algorithm. This study identifies a specific variant that is capable of producing the best solution in a higher order simulated instances and complex realistic scenario such as an automotive electronic parts supply chain in Malaysia.