Results Versus Needs in Electromechanical Reliability 640577

Emphasis on electronic component reliability has left electromechanical component reliability in comparative neglect, despite the fact that, even in the most advanced electronic systems, there is a large population of electromechanical components. Detailed study of the problem has shown that failures of these components do not fit any single pattern of general distribution theory and that generic failure rates are not valid. The failures can only be explained by the diversity of environments within electronic systems to which such components are exposed. “Figure of Hazard” ratings were derived which generalize reliability prediction based on basic failure mechanisms such as fatigue, stress, wear, etc. Further research showed that these ratings differ significantly for failure mechanisms noted during corrective maintenance and those during preventive maintenance. The conclusions of the study reported on are fourfold: (1) the need to relate reliability to underlying physical phenomena, (2) the need to examine the “micro-environments” of components rather than the “macro-environments” of equipment, (3) the usefulness of “Figure of Hazard” ratings in theoretical and applied design research, and (4) the need to shift as much failure incidence as possible from the corrective to the preventive maintenance phase.