ITC/USA 2007 Conference Proceedings / The Forty-Third Annual International Telemetering Conference and Technical Exhibition / October 22-25, 2007 / Riviera Hotel & Convention Center, Las Vegas, Nevada / While any vehicle that is typically part of a flight test campaign is heavily instrumented to validate its performance, long term vehicle health monitoring is performed by a significantly reduced number of sensors due to a number of issues including cost, weight and maintainability. The development and deployment of smart sensor buses has reached a time in which they can be integrated into a larger data acquisition system environment. The benefits of these types of buses include a significant reduction in the amount of wiring and overall system complexity by placing the appropriate signal conditioners close to their respective sensors and providing data back over a common bus, that also provides a single power source. The use of a smart-sensor data collection bus, such as IntelliBus™1 or IEEE-1451, along with the continued miniaturization of signal conditioning devices, leads to the interesting possibility of permanently embedding data collection capabilities within a vehicle after the initial flight test effort has completed, providing long-term health-monitoring and diagnostic functionality that is not available today. This paper will discuss the system considerations and the benefits of a smart sensor based system and how pieces can be transitioned from flight qualification to long-term vehicle health monitoring in production vehicles.

Pour répondre aux enjeux de la longue vie opérationnelle des systèmes et de l’incertitude de l’environnement, la résilience complète la sûreté de fonctionnement pour prendre en compte les situations imprévisibles, sans précédent : l’objectif est de comprendre la situation pour éviter un accident. La qualité de l’interaction homme-machine est essentielle pour atteindre cet objectif. L’état de l’art présente la résilience des systèmes sociotechniques comme complémentaire à la sécurité. Mettre en œuvre la résilience affecte tant l’architecture système que les processus d’ingénierie système. Enfin, elle affecte aussi l’interaction homme-machine, tant son processus de conception centrée utilisateur, ses modèles utilisateur (persona), que ses modèles d’architecture. Nous avons créé le patron de conception « surveiller et alerter » appliqué à la fonction « éviter » de la résilience, pour donner aux opérateurs la capacité de comprendre la dynamique du système, le conduire à vue face à des situations imprévisibles, sans précédent afin d’éviter la survenue d’un accident. La proposition comprend aussi des processus à mettre en œuvre pour contribuer à la résilience d’un système critique à longue durée de vie. L’application au domaine ferroviaire s’appuie sur l’analyse de rapports d’enquête technique d’accidents. Elle se décline sur le patron de conception « surveiller et alerter » et sur le persona, in fine pour proposer des améliorations des interfaces utilisateur. Des perspectives de recherche complètent le mémoire. / The long operational lifecycle of systems and the uncertainty of the environment are a great challenge to engineers. Resilience enhances reliability and safety to take into account the unforeseeable situations, without precedent. The goal is to understand the situation to avoid an accident. The quality of the human-machine interaction is the key issue to achieve this goal. The state of the art explains that sociotechnical systems resilience completes safety approach. Implementaing resilience impacts both system architecture and systems engineering processes. At the end, implementing resilience impacts human-computer interaction, user centred design as well as architecture models. We created the design pattern “to monitor and alert” applied to the function “to avoid” of the resilience. Its goal is to give to the operators the capacity to understand the dynamics of the system, to control at sight vis-a-vis unforeseeable situations, in order to avoid an accident. The proposal contents the processes to be implemented to contribute to the resilience of long lifecycle critical systems. The application to the railway domain is based on the analysis of three accident technical reports. It is declined, on the processes to be implemented to contribute to the resilience of a system, on the design pattern “to monitor and alert” for the architecture of a resilient system and to propose improvements of the user interface. Research forecasts supplement the report.