Two - stage stochastic optimization model for the design of a capacitated, multi - product, multi - echelon reverse logistics system: A case study in Turkey

Abstract

Electrical and electronic equipment sector can be classified as one of the fastest growing industrial sectors in the world. Increase in customers’ tendency to have latest technology resulting in an important increase in consumption of products and the reduction in useful product lifecycles. On the other hand, if the increase in waste electrical and electronic equipment (WEEE) cannot be managed properly, it may create a serious threat on human health and ecosystem due to several poisonous and chemical substances and cause important economic loss. By considering all these risks, European Union(EU) introduced a regulation in 2002 to take WEEE system under control. As a candidate for EU membership, Turkey enacted similar regulation by the year of 2012. It is aimed to minimize all possible threats and losses by giving responsibility to producers, distributors and municipalities for WEEE collecting and recycling activities. The regulation is relatively new in Turkey. In addition, customer awareness about the importance of WEEE system is not high enough. Thus, there is an important uncertainty in total return amounts. In this study, we aim to create an implementable and efficient reverse logistics(RL) network design tool for the WEEE collection system in Turkey with the presence of multiple product types, uncertain return quantities and capacity constraints for recycling activities. The methodology is based on RL network design using a two stage stochastic mixed integer linear programming with the objective of minimizing the total cost of the system for various return scenarios. The proposed model is tested by considering the collection and recycling system of a major white goods producer in Turkey. Since the proposed model is an NP hard problem, sample average approximation is used as a heuristic method for the proposed model under different number of return scenarios.