Predict the unpredictable Benefits and limits of machine data analytics and component health prediction

Vehicle uptime during the lifecycle largely depends on the right component design as well as timely maintenance and service. From an engineering perspective, the basis for replacing guesswork by prediction is precise knowledge of both the load cycles and wear conditions of the vehicle and its components. Monitoring load cycles, trends, deviations and structural borne sound and their correlation to component fatigue data are today's tools to gather instantaneous insights to machine health beyond common tracking of machine hours. To collect and analyze these data automatically along with the necessary modelling of product knowledge becomes much easier within a connected solution. Moreover, it is applicable to entire fleets instead of single "wired" vehicles. In this contribution, we highlight engineering procedures and algorithms for connectivity supported analytics to predict component health and residual lifetime. Besides lifetime prediction, these comprehensively processed smart data sets also deliver engineering-ready real world specs for next generation vehicles to R&D departments.