Designing dynamic reverse logistics network for post-sale service

Abstract

The paper addresses the problem of designing a multi-country production-distribution network that also provides services such as repairs and remanufacturing. The proposed work concentrates primarily on post-sale service provided by the firm under warranty returns. The proposed model assumes that existing warehouses can also serve as collection centres or repair centres for reverse logistics. In addition, the model also explores the possibility of establishing a new facility. Hybrid facilities are considered because of their huge cost-cutting potential due to equipment sharing and space sharing. The capacity of hybrid facilities can be expanded to a predefined limit to process returned products without hampering forward logistics operations. However, if a product cannot be repaired at the warehouse, it is transported to the plant for remanufacturing. The model optimizes the overall configuration and operation cost of the production-distribution network. The production-distribution model developed in the paper is a mixed-integer nonlinear program (MINLP) that is later transformed to a mixed-integer linear program to reduce the solution time. The usefulness of the model is illustrated using a randomly generated dataset. The model identifies (a) the optimal locations/allocations of the existing/new facilities, (b) the distribution of returned products for refurbishing and remanufacturing, and (c) the capacity expansion of the existing plants and warehouses to facilitate remanufacturing and repair services.

An Insight into Reverse Logistics with a Focus on Collection Systems

“…The center location/allocation decision is usually represented by binary variables to determine whether a facility is opened in a specific region or at a specific moment to expand the capacity of the center. Mishra and Singh [67] determined a hybrid warehouse that can be used for collection and repair. Reddy et al [68] considered the opening of an inspection center at a specific time to incorporate the holding cost.…”

“…[79][80][81] The constraint of capacity is the key characteristic of the collection system. Capacity not only considers the facility [67,69], but it also includes the capacity of the vehicles [68]. An example of capacity constraint is ensuring that enough inventory spaces are available during each time period, whereas the objective function minimizes the inventory through its cost.…”

“…The flow balance constraints are constraints to ensure balance in the flow of product entering and leaving the adjacent components of a collection system. The products leaving the collection system can be delivered for remanufacturing or disposal [67][68][69].…”

“…The time constraint considers constraints to the period of inventory value. There are three types of the periods: the single period [69] where the models are static because no subsequent data are required, two period [78], or multiperiod [67,70,72,76]. The difference in the last two types is the time horizon where some researchers reported that it is sufficient to test the model by using only two periods instead of utilizing multiple periods.…”

“…The product type constraints consist of single product [68,69] or multiproduct [67] constraints. However, the multiproduct model is not necessarily represented by different products because these products can move in the same network link simultaneously [82].…”