Title :
Prediction of product failure rate due to event-related failure mechanisms
Author :
Lin, Don L. ; Welsher, Teny L.
Author_Institution :
Lucent Technol., AT&T Bell Labs., Princeton, NJ, USA
Abstract :
Individual events in the use environment such as accidentally dropping a cellular phone or zapping it with human body ESD (electrostatic discharge) are getting more frequent as electronic products are becoming more portable. Monte Carlo simulations of the stress distribution offered by the environment and the product strength distribution are used to derive the infant mortality (early failure) curve. Fitting the slope of the infant mortality curve is an indicator of how far apart the two distributions are and the frequency of individual events. Two new metrics, SIM (severity of infant mortality) and D5%, to track infant mortality are proposed. The process to set test-based reliability requirements for achieving a given field return goal is also illustrated
Keywords :
Monte Carlo methods; curve fitting; fault trees; reliability theory; Monte Carlo simulations; early failure curve; electronic products; event-related failure mechanisms; field return goal; infant mortality curve; product failure rate prediction; product strength distribution; severity of infant mortality; stress distribution; test-based reliability requirements; Cellular phones; Circuits; Consumer electronics; Curve fitting; Electrostatic discharge; Failure analysis; Frequency; Hazards; Humans; Predictive models; Safety; Stress; Testing; Time to market;
Conference_Titel :
Reliability and Maintainability Symposium, 1998. Proceedings., Annual
Conference_Location :
Anaheim, CA
Print_ISBN :
0-7803-4362-X
DOI :
10.1109/RAMS.1998.653802