Commande optimale stochastique des systèmes manufacturiers en boucle fermée sous contrôle des émissions de gaz à effet de serre (GES)

Abstract

Over the years, investments by governmental and non-governmental organizations have been heavily focused on addressing the threats posed by global climate change. In the example of North American countries, the emissions trading model of the Western Climate Initiative (WCI) program allows for action to reduce greenhouse gas (GHG) emissions. In addition, for a responsible consumption of our natural resources, the concept of the circular economy is attracting a lot of attention to address its environmental concerns. In Canada, through the major orientations of the federal sustainable development strategy, its deployment is increasingly growing, especially in the manufacturing sector through reverse logistics. In order to preserve our environment, managers of manufacturing industries must appropriate these environmental challenges in their production practices without losing sight of the economic aspect. It is within this framework that the work of this thesis aims at integrating the control of GHG emissions and reverse logistics in the control of manufacturing systems in a dynamic stochastic context. The manufacturing systems studied are subject to random phenomena such as machines breakdowns and repairs and generate GHG emissions harmful to the environment. The control problem becomes very complex with the consideration of all these aspects. The contributions of this paper are presented in three parts. In the first part, we consider that the manufacturing system consists of a machine producing a single type of product and assume that its failure rate and emission index are constant. The study is to develop the production feedback control policy including the emission level in the decision making in order to minimize the total cost. In addition to the cost of inventorying finished products, shortage in the total cost, we also took into account the cost of emissions exceeding the maximum limit imposed by the competent authorities. However, in industrial practice the system deteriorates, so affects its reliability and emission rate. The repairs carried out following the occurrence of breakdowns are minimal, so the sending of preventive maintenance is necessary. The second part allows us to integrate in the first one the progressive degradation of the manufacturing system during its operation affecting its reliability and its emission rate. We assume that the failure rate and the emission index depend on the age of the machine. Due to the minimal repairs and the aging effect, it is interesting to apply the preventive maintenance activity. Backward control policies (production and preventive maintenance) are developed in this section to minimize the total cost. In addition to the costs of inventorying finished products, shortages, the cost of emissions exceeding the maximum limit imposed by the competent authorities in the total cost, we have integrated the cost of corrective and preventive maintenance. The last part is the study of the production planning problem in a reverse logistics context under control of its GHG emissions. Backward production control policies for a hybrid closedloop manufacturing/remanufacturing system are developed to minimize the total cost. Since the manufacturing machine produces from raw material and the remanufacturing machine allows the remanufacturing of used products recovered from the market, we have integrated the cost of manufacturing and remanufacturing in the total cost. In this work, we used for the modeling, on the one hand, homogeneous Markovian processes and, on the other hand, non-homogeneous semi-Markovian processes where the dynamics of the machine was affected by its history. A numerical resolution of the optimality conditions, described by the Hamilton-Jacobi-Bellman (HJB) equations, led to the solution of the studied problem. To validate the structure of our obtained control policies, a sensitivity analysis was performed. We continued with a comparative study in order to show the advantage of our proposed policies to those in the literature. Finally, the work ends with a managerial implication to facilitate the implementation of our results. This thesis demonstrates the importance and the possibility of integrating these environmental requirements in terms of GHG emission reductions and the recovery of used products in the control of manufacturing systems in order to preserve our environment. Under reasonable assumptions, the results obtained can benefit a variety of industrial sectors such as automotive, rail and aircraft assembly lines, construction machinery manufacturing plants, material handling equipment and mining.

Date	13 Dec 2021
Original language	French
Awarding Institution	École de technologie supérieure
Supervisor	Jean-Pierre Kenné (Supervisor) & Ali Gharbi (Co-supervisor)