This paper introduces an economic disposal and lot-sizing problem (EDLSP) for perishable inventories, it is a new version of an economic lot-sizing problem (ELSP). The classic ELSP aims to determine the production-inventory strategy such that the sum of holding, production, and setup costs is minimized. In the proposed EDLSP, the optimal disposal strategy, as well as the optimal production-inventory strategy, is determined for perishable inventories. Disposal strategy determines how many and when corrupt items should be removed from the stock. In the new model, the unit holding cost depends on the volume of the corrupt inventories that exist in the stock. To closely reflect reality, the model considers fixed and variable disposal costs that represent the costs of removing the corrupt inventories from the warehouse. As a generalized form of the classic lot-sizing problem with the setup cost, the EDLSP is an NP-complete problem. Therefore, two meta-heuristic algorithms, namely Binary Dragonfly Algorithm (BDA) and Genetic Algorithm (GA), are proposed to solve the model. The Taguchi method is also applied to calibrate the meta-heuristics. Then, the performance of these algorithms is evaluated on a set of test problems. The obtained results indicate that both meta-heuristics have a good performance in solving small-sized problems. However, the GA has a better performance than the BDA for medium- and large-sized problems. Furthermore, a sensitivity analysis of key parameters is done to determine the model specifications and to present some managerial insights.
Bibliografisk notePublisher Copyright:
© 2021, The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature.