Comparison of single-echelon vs. multi-echelon inventory systems using multi-objective stochastic modelling

Abstract

Recent advancements in inventory management have identified need for investigating benefits of single-echelon and multi-echelon systems. Single-echelon inventory control problems focus on determining the appropriate level of inventory for an individual unit within the supply chain network, while multi-echelon inventory optimisation takes a holistic approach by focusing on the correct levels of inventory across the entire network. This paper presents a stochastic modelling approach for multi-tier supply chains with multiple inventory items. The approach will be used to compare behaviour of single-echelon vs. multi-echelon inventory systems. A multi-objective optimisation method is used to optimise model's behaviour with two conflicting objectives: minimising average inventory across the supply chain and maximising overall fill rate. The results show that single-echelon systems result in higher inventory levels and lower fill rates, while multi-echelon systems result in lower inventory levels while maintaining higher fill rates for the entire supply chain network. The methods are demonstrated on a sample network database.