Stochastic Multi-Objectives Supply Chain Optimization with Forecasting Partial Backordering Rate: A Novel Hybrid Method of Meta Goal Programming and Evolutionary Algorithms

Abstract

This study proposes a model for a multi-objective, multi-buyer, multi-vendor, multi-product and multi-constraint supply chain. The buyers’ demand rates are stochastic variables with known probability distribution functions. The classical ([Formula: see text]) inventory control system is used to manage the inventories of all buyers where the lead times are production rate dependent. Shortage is permitted and is partially backordered where the partial back ordering rate is forecasted. The model is considered as a multi-objective integer nonlinear programming problem including cost, service level and lead time objectives and using a novel hybrid method, a hybrid of Meta Goal Programming (MGP) and Firefly Algorithm (FA) are solved. Numerical examples are given to illustrate the proposed method in the study. The results of the study are compared to other hybrid methods of Meta Goal Programming with other evolutionary algorithms such as Bees Algorithms (BA), Particle Swarm Optimization (PSO), Genetic Algorithm (GA) and Simulated Annealing (SA).