A contribution for the mixed no-idle flowshop scheduling problem with sequence-dependent setup times: analysis and solutions procedures

• Ano de defesa: 2019
• Autor(a) principal: Rossi, Fernando Luís
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: eng
• Instituição de defesa: Biblioteca Digitais de Teses e Dissertações da USP
• Palavras-chave em Português: Flowshop; Heurísticas; Heuristics; No-idle; Setup Times; Tempos de Preparação
• Link de acesso: https://www.teses.usp.br/teses/disponiveis/18/18156/tde-20052020-110437/

In this work the mixed no-idle permutation flowshop with sequence-depdent setup times scheduling problem is approached. This production environment has not yet been studied in the literature, despite being present in the dynamics of production systems. In the no-idle flowshop environment, the machines cannot be idle and all jobs are processed uninterruptedly. Generally, the reasons are associated with economic or technological factors, where a stationary machine directly influences the performance of the production system. The no-idle flowshop is present in the manufacturing of fiberglass, production of integrated circuits, in steelworks, among others. However, assuming that all machines cannot be idle is often unrealistic. A more realistic situation would consider a mixed environment, where only a few machines perform the jobs uninterruptedly, while the other allow idleness. In this case, the environment is called mixed no-idle flowshop. In the problem extension studied in this work, setup times are considered on machines where idleness is allowed. This is the first work that addresses the mixed no-idle flow shop scheduling problem with setup times. In this Thesis, efficient heuristic methods for solving the mixed no-idle flowshop with setup times scheduling problem are proposed. To demonstrate the performance of the new methods, extensive comparisons with state-of-the-art methods from literature are performed. The results show that the proposed heuristics provide high quality solutions with computational efficiency, outperforming the methods from the literature.