Reliability Prediction for Automotive Electronics Based on Field Return Data
This document should be used as guidance for non-handbook based reliability predictions conducted on automotive electronics products. It presents a method that utilizes warranty and field repair data to calculate the failure rates of individual electronic components and predict the reliability of the entire electronic system. It assumes that the user has access to a database containing field return data with classification of components, times to failure, and a total number of components operating in the field.
In early design activities (typically before the hardware is built), a reliability prediction is often required for the electronic components and systems in order to assess their future reliability and in many cases to meet customer specifications. Those specifications may include the allocated reliability for a particular electronic unit and in the cases of functional safety products to meet the ASIL (Automotive Safety and Integrity Level) requirement specified by the international functional safety standard ISO 26262.
This Recommended Practice (RP) document will provide guidance on performing reliability predictions for automotive electronic products utilizing field return data or any other types of failure data available to an automotive electronics supplier. This document will cover the possible sources of data, types of the data required, ways to collect it, and the methodology of how to process these data to calculate the failure rates and the expected reliability. This document will also include a case study based on the data obtained by Delphi Electronics & Safety, a Tier 1 automotive supplier to illustrate the process of reliability prediction based on field return data.