Research On Inventory-Routing Problem Considering Products Return

With the development of technology and globalization of economy, the key pointof enterprise competition changes from the traditional quality and function to supplychain. Inventory strategy and routing management are two critical parts of supplychain, the combined optimization of them becomes a hot topic in recent years alongwith the vendor management inventory’s blossom in order to deal with the trade-offbetween inventory and routing management. Another, the products return is more andmore generic with the prosperity of E-commerce. Thus, the combined optimization ofinventory and routingproblem (IRP) considering products return in this paper isessential in both theory and application.This paper focuses on two scenarios, the single-period IRP with discrete randomdemand and the multi-period IRP with continuous random demand based on recentresearch status quo in IRP. In the study of the single-period IRP, products unused needto be recycled. This model is a multi-objective optimization problem aimed atminimizing the total cost which includes inventory cost, distribution cost, recyclingcost and minimizing the average delivery time. Then we establish a multi-objectivegenetic algorithm to solve the model whose chromosome has three stages that makethe output clarity. Learn from Solomon’s VRP benchmarks, we set the IRP benchmarksto verify the feasibility of the model and algorithm. And contrast the results of threeorder policy they are the optimal order quantity, sub-optimal order quantity and thissingle-period IRP model.In the study of the multi-period IRP, each period will have some return requires tobe recycled together with distribution. In the model, retailers choose the dynamicinventory policy while the ideas of distribution costs sharing is used to calculate eachretailer’s distribution cost and the average distribution time is also an importantqualification. Then IRP benchmarks used to verify the feasibility of the model andalgorithm. We found that for a single enterprise returns recovery is the way to enhancethe quality of service but not with the economy after contrasting the results betweenrecycling return and non-recycling return but into the current demand. In this paper, we provide a way to processing the return in the supply chain andverify the rationality and economy of this approach. However, this study only focuseson a one-to-many system with a single product and two levels. Future research canfocus on the many-to-many system with multi-products and three levels. One of theinnovation in this paper is using average distribution time to be the main indicators ofthe level of service, more methods will be used in the further study. Further study alsocan pay more attention to more return approach and different inventory policy.