Advances in Hardware Reliability Prediction

Business and engineering goals normally require that reliability predictions be provided quickly, consistently, systematically, accurately, and with known precision. Many models (e.g., Telcordia 1997) have been developed and used that readily satisfy the first three of these requirements. The latter two, however, have provided a rich source of discussions and even controversy within the profession. Our purpose in this paper is to describe a new, practical way of making visible the precision of hardware reliability predictions for subassemblies of electronic systems. We do not specifically address the accuracy issue in this paper as it raises fundamental non-mathematical issues whose discussion more properly belongs in forums of the reliability engineering profession.