A multi-start metaheuristic approach for multi-objective car resequencing problem in just-in-time assembly lines : a real-world
Visualizar/ Abrir
Data
2025-11-11Autor
Frazon, Aline
Kleina, Mariana
Choueiri, Alexandre Checoli
Cunha, Henrique
Oliveira, João Henrique Garbin de
Universidade Federal do Paraná. Setor de Tecnologia. Curso de Graduação em Engenharia de Produção
Metadata
Mostrar registro completoResumo
Abstract : This paper addresses the problem of car resequencing (CRP) in mixed automotive assembly lines, aiming to minimize the impacts of supplier delays in companies with a Just-In-Time (JIT) system. In this scenario, when there is a shortage of parts from suppliers, the cars that use these parts must be resequenced to avoid production stoppages. However, the resequencing must comply with several constraints related to production leveling, safety, operational effort, among others. Thus, using a multi-objective approach, the study proposes a multi-objective multi-start meta-heuristic that is unprecedented in the literature. The solution was implemented in Visual Basic for Applications (VBA) Excel to facilitate integration into the planning routine of the company studied. Tests demonstrated average reductions of 90% in resequencing time and up to 47% in delayed positions, significantly improving the performance of the tool compared to the manual process. The tool automates Pareto dominance analysis, offering optimized solutions and support for decision making. In addition to reducing operational costs and respecting practical limitations, the model contributes to sustainability and innovation goals, aligned with the Sustainable Development Goals (SDGs)