Skip to Main content Skip to Navigation

Apports des connaissances métiers pour la conception d'un système multi-capteurs de tri qualité Billon

Abstract : This study takes place in the frame of a university/industry convention between the CRAN and Siat-Braun sawmill. Sawmill transformation process is divided in two parts. First, the log yard cuts down stem in logs (round timber with standard length) which are sorted out in homogeneous batches by dimensions and quality. Then, the sawing line transforms those logs in end products sorted out by dimensions and quality too. The qualitative sort out of log has a direct impact on sold product quality, so sawmills try to optimize this production step in order to reduce the product downgrading rate. In this goal, Siat-Braun sawmill invested in a new log yard. This study contributes to the development and the commissioning of this system dedicated to log Quality sort out. The first chapter of this thesis places the project into wood industry then explicates sawmill constraints and needs in term of Quality. A literature review points out different observation systems allowing measurement of wood characteristics in order to make a product and log quality rankings. Observation system chosen by our industrial partner is then presented. The objective of the second chapter is to clarify and defines Log Quality to the new log yard. This quality notion being held by the sorting operator, we suggested extract, formalize and model this knowledge using NIAM-ORM method. This method allowed obtaining two structured ontologies: the Log Quality and the Product Quality which we used in the chapter three in order to underlying 50 links between both qualities usually tackle separately. This study allowed also extracting 34 relevant criteria to specify Log Quality in 9 classes, regarding Product Quality. In addition, knowledge model analysis gives an initial set up to the first commissioning of the Observation system (input criteria, output classes, decision thresholds). This set up is so proven during system exploitation. The first tests on end products and data processing delivered by the sensor are used to validate and refine the set up to improve the log sort out regarding the obtained product quality. So, this study allowed capitalizing Expert knowledge of the project actors and then, obtaining a relevant initial configuration of the sort system for the first use. First analysis allowed improving Log Quality sort out system set up. In a near future, we want to go deeper in analysis of data given by the system in order to improve relevant criteria selection
Document type :
Complete list of metadata

Cited literature [140 references]  Display  Hide  Download
Contributor : Thèses UL Connect in order to contact the contributor
Submitted on : Tuesday, December 11, 2018 - 10:19:37 AM
Last modification on : Saturday, October 16, 2021 - 11:14:15 AM
Long-term archiving on: : Tuesday, March 12, 2019 - 1:12:29 PM


Files produced by the author(s)


  • HAL Id : tel-01750611, version 1



Benjamin Almecija. Apports des connaissances métiers pour la conception d'un système multi-capteurs de tri qualité Billon. Autre. Université de Lorraine, 2013. Français. ⟨NNT : 2013LORR0264⟩. ⟨tel-01750611⟩



Record views


Files downloads