Skip to Main content Skip to Navigation
Theses

L’analyse de la qualité dans un système de fabrication reconfigurable

Abstract : RMS is an active field of research, and its literature has been under discussion for the past twenty years. Several contributions have been offered to advance this field of research. From optimization viewpoint, different studies have been published aiming to analyse the cost, time, modularity, and responsiveness-based performance of RMS. However, a quantitative model to examine the quality of production in RMS has not been considered.Compared to other manufacturing systems, quality is more pertinent and is difficult to analyse in a reconfigurable system because of the following reasons. Firstly, RMS offers the combination of several machine configurations, called production routes, to produce the same product. In other words, there are many possible production routes through which a product may pass in a reconfigurable system. This makes it difficult to analyse the product quality of each route. Secondly, if there is a problematic configuration in any route, it becomes difficult to identify that configuration based on the quality of final product.To address these issues, this research offers a quantitative model to examine the quality of production in the process planning of a reconfigurable manufacturing system. In addition, our aim is to understand how quality-based variation impacts the cost and modularity of RMS. A multi-objective model is formulated which comprises of the objectives of the total cost, the quality decay index, and the modularity effort. A Manufacturing System Design Decomposition (MSDD) framework is used to identify the sources causing quality variation in a reconfigurable manufacturing system. Two models are proposed i.e., Model 1 and Model 2. Model 1 is based on quality variation in a reconfigurable manufacturing system and Model 2 corresponds to a perfect quality based reconfigurable manufacturing system.Since RMS problems are non-polynomial hard (NP-hard), heuristics/evolutionary algorithms are the appropriate set of approaches for solving these problems. A hybrid version of non-sorting genetic algorithm and multi-objective particle swarm optimization is used to solve the multi-objective model. The main findings and implications for practitioners can be summarized as:1.Although RMS is known for its cost-efficiency, it seems that the variation in quality and failed operation units impact its performance. Thus, it is imperative to safeguard it against different sources of variation to perform cost-optimally.2.The presence of quality variation results in a different process plan (Model 1) as opposed to a manufacturing system which does not contain any quality variation (Model 2). Both models performed quite differently in terms of modular needs and number of configurations.3.The presence of quality variation affects the overall efficiency of a process plan. It can be argued that in the absence of variation, even maximum cost solution is more viable than the minimum cost solution in the presence of variation. In addition, less average modular effort is needed by a process plan which is free from defects and variation.4.Practitioners are interested in enhancing the productivity of RMS by minimizing the “reconfiguration” between different operations. The findings suggest that modular efforts and quality variation need to be simultaneously analysed to enhance the overall productivity and efficiency of a process plan.
Complete list of metadata

https://pastel.archives-ouvertes.fr/tel-03682947
Contributor : ABES STAR :  Contact
Submitted on : Tuesday, May 31, 2022 - 2:06:14 PM
Last modification on : Friday, August 5, 2022 - 2:54:00 PM

File

khan.pdf
Version validated by the jury (STAR)

Identifiers

  • HAL Id : tel-03682947, version 1

Citation

Abdul Salam Khan. L’analyse de la qualité dans un système de fabrication reconfigurable. Génie des procédés. HESAM Université, 2021. Français. ⟨NNT : 2021HESAE054⟩. ⟨tel-03682947⟩

Share

Metrics

Record views

27

Files downloads

32