Reverse Logistics is the flow and management of products, packaging, components, and information from the point of consumption (i.e., the market) to the point of origin (i.e., manufacturers and suppliers). It is a collection of practices similar to those of supply chain management, but in the opposite direction, from downstream to upstream. Reverse logistics is a valuable solution to the hazards jeopardizing the environment, and it involves activities such as reuse, repair, remanufacture, refurbish, reclaim and recycle.Reverse logistics became an established line of research, covering several areas, including inventory control; though, several research gaps still exist, such as: ignoring switching costs between production and remanufacturing processes and learning effects, the assumption that production and remanufacturing processes are of perfect quality, remanufactured products are assumed to be as-good-as new, the assumption that returned products are treated as whole products while ignoring disassembly, collection rate of used items is independent of price and quality, and the assumption that pure remanufacturing and production policies are optimal. These research gaps are addressed in mathematical models to bring reverse logistics optimization closer to reality. Deterministic and stochastic components are considered here with numerical examples and results discussed. The key conclusions are as follows:The inclusion of the first time interval where no remanufacturing/repair exists, results in preventing the overestimation of inventory holding costs in the repairable stock. Assuming production and remanufacturing processes to be perfect, or ignoring learning effects in these processes, might not capture the benefits that product recovery programs are supposed to bring. Although works in the literature assumed pure remanufacturing is mathematically attainable but not feasible, this study shows that the pure remanufacturing case is not valid mathematically, which proves it to be infeasible. It is favourable to compensate customers to settle for remanufactured products instead of new ones. Considering disassembly of returns in the modelling of reverse logistics is proven beneficial. Finally, mixed production and remanufacturing policies are optimal rather than pure ones; and the inclusion of price and quality to determine return and collection rates is crucial.

Reverse Logistics is the flow and management of products, packaging, components, and information from the point of consumption (i.e., the market) to the point of origin (i.e., manufacturers and suppliers). It is a collection of practices similar to those of supply chain management, but in the opposite direction, from downstream to upstream. Reverse logistics is a valuable solution to the hazards jeopardizing the environment, and it involves activities such as reuse, repair, remanufacture, refurbish, reclaim and recycle.Reverse logistics became an established line of research, covering several areas, including inventory control; though, several research gaps still exist, such as: ignoring switching costs between production and remanufacturing processes and learning effects, the assumption that production and remanufacturing processes are of perfect quality, remanufactured products are assumed to be as-good-as new, the assumption that returned products are treated as whole products while ignoring disassembly, collection rate of used items is independent of price and quality, and the assumption that pure remanufacturing and production policies are optimal. These research gaps are addressed in mathematical models to bring reverse logistics optimization closer to reality. Deterministic and stochastic components are considered here with numerical examples and results discussed. The key conclusions are as follows:The inclusion of the first time interval where no remanufacturing/repair exists, results in preventing the overestimation of inventory holding costs in the repairable stock. Assuming production and remanufacturing processes to be perfect, or ignoring learning effects in these processes, might not capture the benefits that product recovery programs are supposed to bring. Although works in the literature assumed pure remanufacturing is mathematically attainable but not feasible, this study shows that the pure remanufacturing case is not valid mathematically, which proves it to be infeasible. It is favourable to compensate customers to settle for remanufactured products instead of new ones. Considering disassembly of returns in the modelling of reverse logistics is proven beneficial. Finally, mixed production and remanufacturing policies are optimal rather than pure ones; and the inclusion of price and quality to determine return and collection rates is crucial.