Intégration de la dynamique de l'humain dans la planification de la production d'un système manufacturier

Abstract

This study addresses the issue of joint planning of production and corrective maintenance in a manufacturing system subject to random failures and repairs. The purpose of this research is to develop a joint policy to control production and corrective maintenance of a manufacturing system by taking in account of the human in the machine repair activities. The performance criteria considered in this work is the total cost of production. This cost consists of inventory and commissioning costs, stock-out and corrective maintenance. The demand for finished products is deterministic and defined by a constant rate. The models developed in this study are an extension of the classical model of Akella and Kumar (1986) applied to a system consisting of a machine producing one type of product (M1P1). In fact, those models consider the degradation of the repair technician’s performance (increased repair time) in the planning process of the production and corrective maintenance. The proposed approach is based on stochastic optimal control applied to a production system modelled by a Markov process continuous-time dynamic programming is used by a discounted cost over an infinite horizon. The optimum conditions obtained are described by partial differential equations coupled in the form of the Hamilton-Jacobi-Bellman (HJB) equations. These equations are difficult to solve. Therefore, we propose a numerical solution using the approach of Kushner. This numerical solution has allowed us to identify joint control laws (production rate of the machine and replacing the repair technician) considered the production system. A detailed analysis of the results and sensitivity analysis provided a description of the production plan (support tool for decision) and to validate the proposed models for different scenarios associated with degradation of the performance of the repair technician.

Date	9 Feb 2017
Original language	French
Awarding Institution	École de technologie supérieure
Supervisor	Jean-Pierre Kenné (Supervisor)