Soutenance de thèse : Yenny Alexandra PAREDES-ASTUDILLO

Flowshop scheduling problem including learning and fatigue effects: theoretical contributions and case study

Doctorante : Yenny Alexandra PAREDES-ASTUDILLO

Laboratoire INSA : DISP

Ecole doctorale : ED512 : InfoMaths

Despite the rise of automation, hand-intensive production systems are still in use due to the need for workers to be extremely flexible and precise in completing certain tasks. Interest is generated by the impact of learning and fatigue on manual task productivity. The focus of this study is on a flowshop scheduling problem (FSSP) that considers the effects of learning and fatigue. Firstly, a literature review was conducted on the scheduling problem with learning and deterioration effect. After that, a theoretical approach was used to address the problem of FSSP with learning effect. Mathematical models that minimize the makespan were presented. Exact methods and heuristics were proposed for solving the problem in small and large instances. As fatigue is a type of deterioration, a multi-agent model was proposed to validate the integration of muscular fatigue into the FSSP by minimizing the total fatigue dose. Finally, the framework was validated by a case study application that took place in a manual picking line. The modeling of learning and fatigue effects and the computation of model parameters from real data were discussed. A bi-objective approach was proposed to minimize both makespan and total fatigue dose simultaneously. Break policies are recommended depending on the company's needs and the objective to prioritize. The aim of this work is to inspire future work that is interested in addressing operations research problems with a responsible incorporation of human factors.