A decomposition heuristic to solve the two-stage lot sizing and scheduling problem with temporal cleaning

Autor(es)

Toscano, Alyne ^[1] ; Ferreira, Deisemara ^[2] ; Morabito, Reinaldo ^[3]

Afiliação do(s) autor(es)

^[1] Univ Fed Triangulo Mineiro, Dept Prod Engn, Uberaba, MG - Brazil

^[2] Univ Fed Sao Carlos, Dept Phys Chem & Math, Sorocaba, SP - Brazil

^[3] Univ Fed Sao Carlos, Dept Prod Engn, Sao Carlos, SP - Brazil

• Tipo de documento: Artigo Científico
• Fonte: FLEXIBLE SERVICES AND MANUFACTURING JOURNAL; v. 31, n. 1, p. 142-173, MAR 2019.
• Citações Web of Science: 0

Resumo

This paper presents a study on the fruit-based beverage lot sizing and scheduling problem. This is a two-stage lot-scheduling problem with some features, such as the presence of buffers between the production stages, mandatory temporal cleanings for the preparation tanks (stage I) and production lines (stage II), and production synchronization between the production stages. To solve this problem, we propose two variations of a two-phase heuristic algorithm by devising a production stage decomposition scheme coupled with a constructive procedure. In the first phase, a one-stage lot sizing is solved by a mixed integer programming model, and in the second phase, scheduling and the synchrony between the stages are carried out using a constructive algorithm. The two variations of the heuristic are: Stage I Heuristics (S1H) and Stage II Heuristics (S2H). In S1H, production stage I is solved by a mathematical model and in S2H, production stage II is solved by another mathematical model. Data from a typical beverage company of a large worldwide brand was used in the computational tests. The results showed that the heuristics are able to find competitive solutions very quickly and that the S2H heuristic performs better than S1H. (AU)

• Processo FAPESP: 10/10133-0 - Problemas de corte, empacotamento, dimensionamento de lotes e programação da produção, e suas integrações em contextos industriais e logísticos
• Beneficiário: Reinaldo Morabito Neto
• Modalidade de apoio: Auxílio à Pesquisa - Temático