The order fulfillment problem in the textile industry: MIP models for service-levels constraints and integrated production-distribution problems

• Ano de defesa: 2024
• Autor(a) principal: Alves, Giovanna Abreu
• Orientador(a): Tavares Neto, Roberto Fernandes
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: eng
• Instituição de defesa: Universidade Federal de São Carlos; Câmpus São Carlos
• Programa de Pós-Graduação: Programa de Pós-Graduação em Engenharia de Produção - PPGEP
• Palavras-chave em Inglês: Mixed integer programming; Order fulfillment; Lot-sizing problems; Integrated production and distribution; Service-level constraints
• Área do conhecimento CNPq: ENGENHARIAS::ENGENHARIA DE PRODUCAO::PESQUISA OPERACIONAL
• Link de acesso: https://repositorio.ufscar.br/handle/20.500.14289/21353

In this work, we address the production planning problem by focusing on meeting orders in the textile industry and analyzing how mathematical programming can support decision-making. Aiming to characterize the textile industry in São Paulo, Brazil, we performed an exploratory survey with managers of textile industries in the state. The survey highlighted crucial factors for the competitiveness of the companies, such as customer relationship, quick response, and production scheduling. It also shows the lack of management information systems for production planning. Additionally, an extensive literature review further contextualizes these challenges within the scope of textile supply chain management, highlight areas where mathematical programming can be applied to address production planning problems in the textile sector. Based on the identified gaps, we propose two mixed integer programming models for production planning problems focusing on customer order fulfillment. The first model focuses on servicelevel and capacity utilization constraints within capacitated lot-sizing problems, exploring the effects of these constraints on key performance indicators such as delay times, throughput, and capacity utilization. The second model supports integrated production and distribution decisions, demonstrating its advantages over hierarchical models through a comparative analysis. We also proposed an Adaptive Large Neighborhood Search (ALNS) heuristic to solve the integrated model in lower computational times, offering insights into the operational complexities of textile manufacturing.