Comprehensive approach for the reliability prediction of complex systems

In an increasingly competitive environment the products quality is a crucial aspect of its acceptance by the customer. In this context reliability is one of the most significant attributes for customers [1]. That is the reason for the aim to present an approach to predict the reliability of a complex vehicle system. The knowledge about the occurring failures and the knowledge of the reliability in different design stages are used. To show the potential of this approach, the paper presents an application for an Electronic Braking System. In a first step the presented approach shows investigations of the physics of failure based on the corresponding acceleration model. With the knowledge of the acceleration factors, the reliability can be determined for each component of the system for each design stage. If the occurred failure mechanisms are the same for each design stage, the determined reliability of earlier design stages could be used as pre-knowledge as well. In a second step the field reliability of the system can be predicted at the end of the development cycle with the usage of the determined acceleration factors. For the calculation of the system reliability the reliability values of all system components are brought together. In this paper the influence of sample size, stress levels of testing and test duration on reliability characteristics are investigated. Here, an Electronic Braking System serves as the system to be investigated. Accelerated testing is applied, i.e. testing at elevated stress levels. If pre-knowledge is applied and the same test duration is observed, this allows for the conclusion of higher reliability levels. Alternatively the sample size can be reduced compared to the determination of the reliability without the usage of pre-knowledge. It is shown that the presented approach is suitable to determine the reliability of a complex system such as an Electronic Braking System. An important part for the determination of the acceleration fac- or, i.e. the ratio of the lifetime under use conditions to the lifetime under test conditions, is the knowledge of the exact field conditions. Because that is difficult in many cases further investigations are necessary as well. For future analysis the last step could be to use the knowledge of the determined system reliability to adapt the testing for a related system development in the future.