Overview
Read this article from a comprehensive knowledge base, updated and supplemented with articles reviewed by scientific committees.
Read the articleAUTHOR
-
Rachid SABRE: Teacher-researcher - Head of the Laboratory of Mathematics Applied to Computer Science and Statistics MAIS at ENESAD - Member of Institut de Mathématiques de Bourgogne CNRS
INTRODUCTION
In recent years, the agri-food industry, when faced with the challenge of improving or creating a new product, has placed great emphasis on the organization of experimental trials. The traditional step-by-step approach was based on the experimenter's know-how and common sense. This led to numerous trials and a considerable amount of time, sometimes leading to results that were difficult to interpret.
However, customer demands in terms of quality and taste on the one hand, and companies' drive to reduce development costs on the other, require the use of a scientifically rigorous approach: a "design of experiments", also known as an "experimental strategy".
The chocolate industry, for example, needs to produce a number of mixtures, varying the components or simply varying their dosage, in order to obtain a product that meets certain required organoleptic characteristics, such as: melting, fatty, sweet, caramelized, milky taste, persistence of taste, color, etc. The questions posed by a project manager can be summed up in three pertinent questions:
which components should be studied to determine whether they can change the result to achieve the mix that meets the product's expected characteristics?
what are the proportions of these components to be used in this mixture?
how many tests must be performed to obtain the expected answer?
The aim of the design of experiments is to answer these questions by proposing mathematical methods for organizing a reduced number of experimental trials with usable results.
In this work, we first define a design of experiments and then show the scientific and economic advantages of using it. In the following paragraphs, we present the full factorial design, the fractional factorial design and Scheffé models. In order to preserve the confidentiality of certain studies, the data for a few examples are modified or presented only in part.
This document, which presents a few experimental designs, does not claim to cover all existing designs. Additional designs, such as Taguchi's, will be presented in a later issue.
Exclusive to subscribers. 97% yet to be discovered!
You do not have access to this resource.
Click here to request your free trial access!
Already subscribed? Log in!
The Ultimate Scientific and Technical Reference
This article is included in
Food industry
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
Experimental planning in the food industry
Bibliography
Text references
Software
(Non-exhaustive list)
Nemrod, logiciel construit à LPRAI, Université d'Aix-Marseille http://www.nemrodw.com
Sas, Editor Institute
Spad, DECISIA/SPAD https://www.test-and-go.com/fr/ct
Splus,...
Exclusive to subscribers. 97% yet to be discovered!
You do not have access to this resource.
Click here to request your free trial access!
Already subscribed? Log in!
The Ultimate Scientific and Technical Reference