Mike Schwab's idea is a good one for some but not for all PDSE members In some of them replacement probabilities are different, more like those associated with magnetic hysteresis or elastic fatigue: the longer they have gone unreplaced the likelier their replacement becomes.
No single a priori rule is likely to be satisfactory. Data-driven selection of one from among a small number of them is going to be needed. John Gilmore, Ashland, MA 01721 - USA ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
