A method for computing lower confidence limits on system reliability using component failure data with unequal sample sizes

摘要

A method is presented for constructing system reliability usingcomponent failure data when the sample sizes for testing on thecomponent parts differ greatly. The procedure can be applied toweapons systems as easily as subsystems. No assumptions aboutfailure distributions are made. The accuracy of the procedure wasexamined by computer simulations and in this manner the procedurehas demonstrated high accuracy for cases of practical interest.