Assessing the Assumptions Used in Reliability Prediction Modeling

This paper is aimed at improving the reliability of printed circuit board (PCB) in the electronic packaging/manufacturing industry. It summarizes the research done in the area of electronic system reliability, assesses the approaches and assumptions used in the calculation of electronic system failure rates, and proposes a method to calculate the failure rate of the system that does not rely on the assumption of independent failures. The paper starts with an introduction to reliability theory, justifies the role of reliability prediction methods, provides a historical overview, reviews the key models which are currently in use, and presents the proposed algorithm and results. A weighted-rank method is used to find the failure rate dependency (or correlation) between various components