Cette communication traite d’une démarche de co-conception destinée à concevoir un simulateur d’entrainement pour des opérateurs de centrale nucléaire. Cette démarche multi-compétences et diachronique a impliqué la constitution de plusieurs collectifs de travail durant les étapes de ce projet. Dans un contexte de complexité technique, de pluridisciplinarité et de désynchronisation temporelle sur certaines étapes, la connaissance de l’activité et des besoins utilisateurs a constitué un pilier et une ressource fondamentale pour alimenter la conception du simulateur, faire converger les différents acteurs et garder une ligne directrice cohérente. Nous retraçons les ressources mobilisées et quelques points de tension d’ordre temporel de cette expérience pour tirer des enseignements de cette démarche de co-conception.

An unstructured document is made of a collection of potentially important data that are difficult to process through a computer. In the context of exponential data growth, one of the main challenges for companies is to manage, develop and use unstructured data. The collaboration of Famolde and 1A3i, two Europeans companies, has led to the development of new ways to process and compare unstructured data in Famolde's industrial manufacturing process. However, an ergonomic study identified limitations and potential improvements with the solution developed. In this context, we explored the information visualization literature to propose a new prototype designed to facilitate the comparison and classification of information. A multiple views solution based on the initial juxtaposition comparison tool is enhanced with explicit information about the identified differences. Finally we propose experimentation protocol to validate our solution.

Résumé : L’objectif de cette étude était de concevoir un scénario qui propose un cadre écologique et reproductible pour l’évaluation de la charge mentale dans le cadre de l’utilisation d’interface humain-machine en contexte industriel. Afin de répondre à cet objectif, les variations de la charge mentale ont été évaluées lors de la réalisation d’une tâche de maintenance. Les résultats permettent d’observer des variations de charge mentale et des différences de stratégie et de performance chez les opérateurs. L’intérêt principal de ce scénario est de pouvoir observer ces variations au cours des sous-tâches et pas uniquement de façon globale sur l’ensemble de la tâche. Les études futures utilisant ce scénario pourront donc tester différents types d’interfaces et leurs effets sur la charge mentale et la performance des opérateurs.

Abstract : The objective of this study was to design a scenario that proposes an ecological and reproducible framework for the evaluation of mental load in the context of the use of human-machine interface in an industrial context. In order to meet this objective, variations in mental load were evaluated during the performance of a maintenance task. The results show variations in mental load and differences in strategy and performance among operators. The main interest of this scenario is to be able to observe these variations during the subtasks and not only globally on the whole task. Future studies using this scenario will therefore be able to test different types of interfaces and their effects on the mental load and performance of operators.

Compression therapy is a widely used treatment for venous diseases that involves applying pressure to the affected area to improve blood flow and reduce swelling. One of the challenges of compression therapy is that the optimal compression level varies depending on the shape and size of the affected area. To address this challenge, we extracted a set of morphotypes, using 3D body scans of the lower limb. These morphotypes will aid the development of medical compression stockings, therefore they will be useful for industrials, facilitating in the same time the choice of the correct medical device for health practitioners. This can improve the effectiveness of compression therapy and reduce the risk of complications such as skin irritation or discomfort. In this present paper we discuss the unsupervised classification of a 1D shape descriptor and of a 2D shape descriptor to see the discriminatory properties of these two with regards to shape detection and grouping ability. The 1D shape descriptor highlights the lower limb zones, with the most variations and gives the dominant Variables for 4 morphotypes, the 2D shape descriptor breaks down more the shape differences between 4 different morphotypes, both giving valuable information and concentration points for future works, when we will develop a novel 3D shape descriptor.