Overview
Français
ABSTRACT
This article examines, through surface treatment examples, the interest of experimental designs in the search for an optimum through response surface methodology (RSM). Among the experimental design, those of Box-Behnken, Doehlert, central composites, uniform and D-optimal designs may provide valuable assistance. More generally, they provide a continuum of progressive management in a study with a minimum of testing at the lowest cost. Mixture designs are particular experimental designs; they are dedicated to the optimization of mixture formulations whose properties depend on the proportions of the constituents. This approach is illustrated by means of a case study.
Read this article from a comprehensive knowledge base, updated and supplemented with articles reviewed by scientific committees.
Read the articleAUTHORS
-
Amel KAMOUN: Industrial Engineering – Doctorate in Chemistry - Lecturer at the Faculty of Science, University of Sfax (Tunisia)
-
Mohamed Moncef CHAABOUNI: Doctor of Science – Physics - Professor at the National Engineering School, University of Sfax (Tunisia)
-
Hassine Ferid AYEDI: Doctor of Science – Physics - Professor at the National Engineering School, University of Sfax (Tunisia)
INTRODUCTION
In
The steps involved in this procedure, known as the "Response Surface Methodology" (RSM), can be summarized as follows:
project audit. Its aim is to set the objectives of the approach and identify the factors influencing the process and/or product, with the active participation of the entire project group. This starting point is crucial: many approaches fail for lack of a good initial assessment to be carried out with a view to future developments;
experimentation strategy. Following the audit, it is now possible to define a relevant and cost-effective experimentation strategy. To do this, we build the right experimental plan, adapted to the project, enabling us to carry out a minimum number of trials;
plan execution. Rigor and good communication are necessary between the operators and the specialist in experimental strategies;
modeling and analysis of results. Modeling consists in establishing a causal model between the factors and the property of interest, making it possible to define the ranges to be respected in production. This makes it possible to take informed decisions, giving production managers, for example, the optimum value ranges for process operating parameters.
The present article is dedicated to the search for the optimum and to the formulation of TS mixtures through examples drawn from the scientific literature; reasons of industrial confidentiality imposing this choice on us.
In order of appearance, the plans for :
Box-Behnken ;
Doehlert;
centered composites ;
uniforms ;
D-optimal ;
mixtures.
It should be noted that the data kindly supplied by the authors...
The Ultimate Scientific and Technical Reference
This article is included in
Metal treatments
This offer includes:
Knowledge Base
Updated and enriched with articles validated by our scientific committees
Services
A set of exclusive tools to complement the resources
Practical Path
Operational and didactic, to guarantee the acquisition of transversal skills
Doc & Quiz
Interactive articles with quizzes, for constructive reading
Design of experiments and surface treatments
Finding an optimum with MSR
Bibliography
Software tools
Non-exhaustive list
• CORICO, distributed by CORYENT conseil
http://www.coryent.com/mlesty.html
• DESIGN EXPERT, distributed by Stat-Ease
http://www.statease.com/software.html
• DOE++, distributed...
The Ultimate Scientific and Technical Reference
