Accelerated Reliability / Durability and Field Test Data: Collection and Analysis for Decision Making. 2012-01-0204
Accelerated Reliability/Durability testing (ART/ADT) is an extremely important and quite complicated problem of reliability theory and practice. It is known to be today the most efficient reliability estimation of equipment and systems. The main problems reliability engineer withstands with in connection with the accelerated tests are twofold:
How to be sure that we can successfully and appropriately apply the acceleration to the system under observation (SUO)
How to handle statistics obtained when the situation is not “clear-cut”
This paper is dealing with each of the above questions trying to give a comprehensive organizational and statistical framework for an accelerated Reliability test planning, tracking and decision making. Accurate prediction and control of reliability and durability plays a very important role in the profitability of a product. Service costs for products within the warranty period or under a service contract are a major expense and a significant pricing factor. Proper spare part stocking and support personnel hiring and training also depend upon good reliability fallout predictions. On the other hand, missing essential reliability targets may raise contractual penalties, tangible and intangible costs in the future.
The paper concludes the united accelerated reliability testing approach for collecting accelerated/field reliability data for a class of products without sequence effect, so called Products Invariant to Order of Load Condition (PIOLC). Then the paper suggests a universal numerical data analysis procedure for decision making about the distribution parameters predictable values, based on Maximum Likelihood Evaluation method and Cross-Entropy optimization approach.