Skip to Main content Skip to Navigation
Master Thesis

Prédiction des fibres des matières premières utilisées en nutrition animale

Abstract : Fiber is defined by the method applied for its analysis: crude fiber (CB), neutral detergent fiber (NDF), acid detergent fiber (ADF), acid detergent lignin (ADL), insoluble and soluble fiber (FI and FS). These methods of analysis are long and expensive whereas it is important to know the fiber level of a raw material before formulation rations. The purpose of this study is to define equations to predict the different fiber levels (NDF, ADF, ADL, FI and FS) of raw materials used in animal feeding. A total of 276 samples representing 40 different raw materials was analysed for fiber levels and chemical constituents (moisture, protein, fat, starch, sugars, CB, minerals). Statistical analysis of the data was performed thanks to the software R. Raw materials have been divided into 3 groups: the rich in fiber, the rich in protein and the rich in energy. For each group, correlations have been calculated to select relevant models of linear regression. These models have been performed for each raw material and the best model according to the higher ratio performance deviance (RPD) was chosen like equation to predict the fiber level of the raw material. The results of this study show it is possible to predict NDF, ADF and FI for most of the raw materials, but it is difficult for ADL and FS. However, chosen equations are not strong due to the low number of samples by raw material. So, this study ended by the development of a statistical program which performs new equations with additional data in the future.
Document type :
Master Thesis
Complete list of metadata

Cited literature [23 references]  Display  Hide  Download
Contributor : Agrocampus Ouest <>
Submitted on : Wednesday, October 21, 2015 - 2:50:25 PM
Last modification on : Monday, July 6, 2020 - 3:38:21 PM
Long-term archiving on: : Saturday, January 23, 2016 - 12:20:57 AM


 Restricted access
To satisfy the distribution rights of the publisher, the document is embargoed until : 2025-10-27

Please log in to resquest access to the document


  • HAL Id : dumas-01218628, version 1



Manon Devine. Prédiction des fibres des matières premières utilisées en nutrition animale. Sciences du Vivant [q-bio]. 2015. ⟨dumas-01218628⟩



Record views