Watch Now 25:42
Reliability Analysis for Non-Repairable Systems
Presenter: Erich Gundlach
Also see Reliability for Repairable Systems
Life Distribution
The presenter demonstrates how to use Life Distribution to answer a number of reliability questions. He shows how to find the number of failures that will occur after a certain point, how long a certain percentage of devices will last without failing, which units are failing due to a particular cause and at what point devices fail due to wearout. He uses device data, and at the final stage, includes warranty data on causes of failure. He shows how to fit a variety of distributions, including competing risk mixtures.
Fit Life by X and Degradation
The presenter uses capacitor data to show how to use accelerating conditions to analyze lifetime events when only one factor is present. He shows how to extrapolate accelerated data down to normal conditions to predict failures. He cautions to make sure that the accelerating conditions do not introduce failures that would not exist under normal conditions. He uses a Gallium Arsenide laser example to predict pseudo failure times, use these times to estimate life distribution and predict the probability of failure under different operating conditions. He describes the difference between destructive and non-destructive degradation and presents options for transforming data.
Resources
- Journal and files
- Applied Reliability by Paul A. Tobias and David Trindade
- Statistical Methods for Reliability Data by William Q. Meeker, Luis A. Escobar