Reliability library using the reliability block diagram method based on field return data

Within the automotive industry context, knowledge of our product reliability is necessary to remain competitive. Capitalizing on our knowledge from field return data to provide reliability information to development teams is a pillar of our R&D strategy, made more accessible today thanks to digital tools. Predictive reliability is a major discipline of research and development activities because it demonstrates the performance of the systems developed for our customers. The main goal of this paper is to illustrate the process of building a system reliability model using field return data, analyzing failures down to the sub-system level and their specific failure effects. The objective is to create and update a sub-system reliability library using the Reliability Block Diagram (RBD) method, which enhances the precision of assessments of system reliability. The stakes of this project are twofold: Demonstrating predictive reliability that aligns with customer targets early in new project development and Capitalizing on a Subsystem reliability library through the analysis and expertise of warranty return parts.
This approach offers several advantages: Knowing component reliability helps identify the most sensitive subsystems for implementing improvement solutions and increasing product reliability. It allows for the updating of Failure Mode, Effects, and Criticality Analyses (FMEA) through both qualitative (failure cause and effect) and quantitative (criticality assessment) studies. Historically, meeting customer reliability requirements often demanded significant support from specialists. The RBD approach simplifies this by providing autonomy to project teams via the component reliability library. Teams can use this library to model the system (e.g., a headlamp) and calculate its predictive reliability. The methodology is founded on field return data (REX), where the warranty returns database (parts failing during the warranty period) is the indispensable tool. The quality of the analysis results is directly dependent on the quality of the recorded data, confirming REX's contribution to reliability mastery.

Keywords: Reliability, FMEA, Reliability block diagram, Field return data, Fault tree analysis.