Stokastik Talep ve Geri Dönüşlü Ekonomik Parti Büyüklüğü Problemi Üzerine Bir Çalışma

Abstract

Due to economic, environmental and legal obligations, the concepts of product reuse and remanufacturing have gained an important dimension both in industry and academia. The fact that remanufactured products are identical to those produced by conventional manufacturing distinguishes remanufacturing from other types of product recovery. Many businesses perform manufacturing and remanufacturing activities together, which causes some difficulties in production and inventory control. One of the most important problems encountered in remanufacturing systems is the adaptation of the classical economic lot sizing problem (ELSP) to the remanufacturing systems. It is known that the economic lot sizing problem with returns (ELSPR), which has a great importance in terms of balancing supply and demand in remanufacturing systems, falls within the class of NP-hard complexity. In other words, it is not possible to obtain the optimal solution of the problem with a computationally efficient algorithm. Furthermore, assuming that both demand and return amounts are deterministic, the applicability of ELSPR to real life conditions is limited. For this reason, the problem in this thesis is addressed by the assumption that the demand and return amounts are stochastic and non-stationary and a computationally efficient heuristic algorithm has been developed for the solution of the problem by adapting the well-known Silver-Meal (SM) heuristic to the remanufacturing systems. It is aimed to determine the inventory plan that will minimize the total expected cost of holding, penalty and set-up costs by deciding when and how much manufacturing and/or remanufacturing will be made for a single product with a finite planning horizon. Finally, a large set of test has been produced in which the various factors that could affect the performance of the algorithm accounted for and the findings of the algorithm's operational effectiveness have been reached.