An integration of multi-objective optimization with the response surface methodology to the DMAIC method: an application in the glass lamination process

• Grant number: 21/10799-2
• Support Opportunities: Regular Research Grants
• Start date: July 01, 2022
• End date: June 30, 2024
• Field of knowledge: Engineering - Production Engineering - Operational Research
• Principal Investigator: Aneirson Francisco da Silva
• Grantee: Aneirson Francisco da Silva
• Host Institution: Faculdade de Engenharia (FEG). Universidade Estadual Paulista (UNESP). Campus de Guaratinguetá. Guaratinguetá , SP, Brazil
• Associated researchers: Fernando Augusto Silva Marins

Abstract

The laminated products meet the strictest international standards for human impact and safety. They have excellent absorption properties, providing a high sound attenuation performance compared to other laminated glasses. Another benefit of glass lamination is improved thermal conductivity due to the higher heat transfer coefficient of the intermediate material. Thus, in many practical situations, it is important to assess the relationships between the factors (or variables) that make up a process, as well as one or more response variables of interest. The main contribution of this study will be to propose a system that combines Multi-objective Optimization with Response Surface Methodology (RSM), agglutination functions and optimization algorithms with the DMAIC method, aiming to optimize a glass cutting process. In this sense, another contribution of this project will be in the application of the process of cutting glass, since there are few works related to this subject. This systematic, will combine the RSM with the functions: Desirability, Modified Desirability and Commitment Programming, being used, the algorithms GRG and Evolutionary for the execution of the optimization. A systematic or framework will be created that shows the integration of these quantitative techniques to the DMAIC method. Thus, showing the steps to use this systematic in a real glass lamination process of a multinational company. However, it is hoped that this framework can be extended to other real problems. In this study, four input variables will be studied: x1 - Speed of the Lapidary Table [m/min]; x2 - Wheel circumferential speed [m/s]; x3 - Over-material [mm] and x4 - Dressing Time [s] and, three answer variables: y1 - Ratio of the total length of the cut by the diameter of the cutting disc [m/mm]; y2 - Dressing frequency [pç] and y3 - Tool compensation [pç]. It is expected that this research can generate scientific contributions, such as, for example, new applications to the literature of RSM, Multi-objective Optimization, agglutinating functions and DMAIC method. Also, it is expected to publish this study in journals with Qualis A1 and, guidelines for doctoral, master's, course completion and scientific initiation papers. Another scientific contribution focuses on the systematics to integrate multi-objective optimization, RSM, binding functions with the DMAIC method, aiming to improve a real glass cutting process. Another contribution of this study is shown in the opportunity for other applications to use this systematic. As a technological contribution to the industrial sector and the dissemination of this systematic so that other researchers and professionals can replicate this study in various real industrial problems. Thus, this is the main regional insertion of this research. However, it is hoped that this systematic can be used by the entire scientific community and professionals who use the DOE and Optimization technique. As a contribution to the research, step by step of this systematic will be shown through virtual classes (non-monetized), which will be disclosed on the Aneirson youtube channel. The practical results that it is intended to achieve, together with the advantages already mentioned, will ratify the contributions of the systematic proposal and its contemporaneity and scientific relevance. (AU)