This paper deals with a production planning and scheduling (PPS) problem. Specifically, in this paper the authors study a multi-level capacitated lot-sizing problem (MLCLSP). They study the impact of not considering scheduling decisions within MLCLSP and the limitation/flaws of well-known previous MLCLSP formulations. They show that most of the solutions reported in the literature obtained by standard MLCLSP with zero lead times are infeasible, and those that consider positive lead times entail significant work-in-process. To this end, the authors present two model formulations. In the first model, based on the batching assumption, the production of a successor item can only start if the complete batch of the predecessor item is finished. In the second model, they relax this constraint and allow simultaneous production of predecessor and successor items. Both formulations are applied to benchmark instances and the differences to the classical model approaches are analyzed. In contrast to MLCLSP with zero lead times, the new model formulations are able to generate always-feasible production plans. Comparing them with MLCLSP with positive lead times illustrates the potential cost savings of 30–40 percent due to reduced total throughput and work-in-process, respectively.
Recensione dell'articolo:(Almeder, Christian; Klabjan, Diego; Traxler, Renate; Almada-Lobo, Bernardo - " Lead time considerations for the multi-level capacitated lot-sizing problem. " - European J. Oper. Res. 241 (2015), no. 3, 727-738.) MR3282287 MathSciNet ISSN 2167-5163
Leonardo Pasini
2015-01-01
Abstract
This paper deals with a production planning and scheduling (PPS) problem. Specifically, in this paper the authors study a multi-level capacitated lot-sizing problem (MLCLSP). They study the impact of not considering scheduling decisions within MLCLSP and the limitation/flaws of well-known previous MLCLSP formulations. They show that most of the solutions reported in the literature obtained by standard MLCLSP with zero lead times are infeasible, and those that consider positive lead times entail significant work-in-process. To this end, the authors present two model formulations. In the first model, based on the batching assumption, the production of a successor item can only start if the complete batch of the predecessor item is finished. In the second model, they relax this constraint and allow simultaneous production of predecessor and successor items. Both formulations are applied to benchmark instances and the differences to the classical model approaches are analyzed. In contrast to MLCLSP with zero lead times, the new model formulations are able to generate always-feasible production plans. Comparing them with MLCLSP with positive lead times illustrates the potential cost savings of 30–40 percent due to reduced total throughput and work-in-process, respectively.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
