AFR: problems of definition, calculation and measurement in acommercial environment

摘要

Most commercial enterprises want a meaningful yet simple methodfor determining product reliability. “Average”,“annual” or “annualized” failure rates areselected because of their apparent simplicity. All involve some sort of“failure rate” averaging over some period of time. All areabbreviated AFR even though there are no accepted industry standarddefinitions for these terms or the processes and conditions for theirmeasurement. If the failure rate is not constant in time, creating anaverage can easily generate meaningless results. A number of businessesuse the term “failure rate” but are in fact calculatingprobabilities or time independent percentages. A basic premise for thispaper is that the true underlying product failure rates are not constantin time. This paper presents a detailed discussion of different wayscommercial industries calculate simple failure rate based metrics. Itidentifies some ambiguities in the definitions and resultantinaccuracies. It will help the reliability pundit understand themathematical considerations, making him more aware of potentialproblems. This paper should help generalists and managers understandthat the AFR's they calculate may be significantly different than thosecalculated by their suppliers and customers. This can result insubstantially different numbers and conclusions. Averaging instantaneousfailure rates should be done only to smooth data collected from adistribution known to have a constant failure rate