A GENETIC ALGORITHM APPROACH FOR DYNAMIC SUPPLIER SELECTION

Abstract

Supplier selection has a great impact on supply chain management. This decision considers many factors such as price, order quantity, quality, and delivery performance. We address a dynamic supplier selection problem (DSSP) which a buyer should procure multiple product from multiple supplier in multiple periods. Furthermore, transportation cost has significant impact in the procurement decision. However, only a few researchers consider transportation cost in their model. This paper proposes a dynamic supplier selection problem considering truckload shipping. A mixed integer non-linear programming (MINLP) model is developed to solve dynamic supplier selection problem. The purpose of model is to assign the best supplier that will be allocated products and to determine the right time to order that can minimize total procurement cost. In addition, constraints such as suppliers’ capacity, truck capacity, inventory balance, service level, and buyer storage are taken into consideration in the model. Due to the complexity of the problem, the formulated problem is NP-hard in nature so a genetic algorithm (GA) is presented to solve dynamic supplier selection problem. Finally numerical example has been solved by the proposed GA and the classical method using Lingo 16. The results illustrate an understandable slight errors in total cost when GA is compared to commonly used classical method.