Optimization of sales plans in closed-loop supply chains with non-stationary consumer behavior

Abstract

In this study, we study the sales plan optimization problem for a manufacturer who can offer both new and remanufactured versions of a product. Demand follows the Bass diffusion process that is widely used in the literature and has been validated with real data for many durable goods. The manufacturer may slow down the diffusion process by delaying the fulfillment of a certain amount of demand in certain time periods. Some of the products sold in earlier periods are returned to the manufacturer in later periods and become available for remanufacturing. The manufacturer may be better off by slowing down the diffusion process to better match the demand for remanufactured products with the used product supply. Customers’ loyalty to the brand (or manufacturer) of the product and the demand for remanufactured products may vary over time. We propose the use of a dynamic programming algorithm to solve this optimization problem for industrial-scaled instances. With this algorithm we conduct numerical experiments and investigate how the optimal sales policies change with respect to the parameters of the diffusion process and consumer behavior. Numerical results reveal that if the customer loyalty to the brand or the demand for remanufactured products decreases over time, the total amount of demand that is immediately met drops in anticipation of generating more revenue from remanufacturing.

Keywords

• Closed-loop supply chains
• Remanufacturing
• Sales planning
• Dynamic programming