Tim Rodgers and I discussed reliability engineering in Fred's "Dare to Know" Podcast (link below).
One of the explanations for the need to change from Reliability Predictions to Empirical Discovery Methods
"Reliability Predictions - Continued Reliance on a Misleading Approach" This most excellent government white paper was given at the 2013 RAMS event should be required reading for all management and reliability engineers focused on traditional reliability predictions. Assumptions of the causes of unreliability in electronics systems continues to misdirect efforts to improve reliability. Download and distribute freely to help eliminate the mythology of the value of reliability predictions for electronics systems.
Grundfos, is an enlightened company when it comes to understanding the fallacy of traditional reliability prediction methods, shared the reasons traditional electronics reliability prediction methods are invalid.
Published in the Nov/Dec 2011 IEEE Design & Test of Computers
"Long-Term Thermal Overstressing of Computers" co-authored with Michael Pecht, Founder and Chairman of the University of Maryland's Center for Advanced Life Cycle Engineering (CALCE)
Significant opportunities exist to reduce costs in the design, manufacture, and operation of systems by using temperatures higher than specified in testing systems' reliability. The authors share the findings and observations of an experimental study in which they subjected operating computers to high steady-state temperatures and thermal cycling well beyond their design specifications. The results suggest that significant cost savings can be realized without compromising reliability .