A multiproduct economic order quantity model with simulated annealing application

Abstract

PurposeThis paper aims to present a set of five models for the economic order quantity problem. Four models solve problems for a single product: incremental discounts with or without backorders and all-unit discounts with or without backorders, and the last model solves problems for the multiproduct case.Design/methodology/approachA basic integer non-linear model with binary variables is presented, and its flexible structure allows for all five models to be utilised with minor modifications for adaptation to individual situations. The multiproduct model takes into consideration the work of Chopra and Meindl (2012), who studied two types of product aggregations: full and adaptive. To find optimal or near-optimal solutions for the multiproduct case, the authors propose a simulated annealing metaheuristic application. Numerical examples are presented to improve the comprehension of each model, and the authors also present the efficiency of the simulated annealing algorithm through an example that aggregates 50 products, each one with different discount schemes and some allowing backorders.FindingsOur model proved to be efficient at finding optimal or near optimal solutions even when confronted with mathematical complexities such as the allowance of backorders and incremental discounts. Originality/valueFinally our model can process a mix of products with different discount schemes at the same time, and the simulated annealing metaheuristics could find optimal or near optimal solutions with very few iterations.