A TPS acceptance procedure calls for inserting 15 faults in a UUT and assuring that all are detected and located properly. If one is missed, five more faults are inserted and tested. This is similar to many TPS acceptance procedures. It seems reasonable and strikes a fair balance between rigor and economics. But what does it means? How good is the TPS that passes the test? Turning the question around, what type of test will ensure that a TPS will find a known percentage of all the possible faults in the UUT? The science of TPS's requires metrics of this nature. The acceptance test procedure described in MIL-STD-2077 addresses this problem, but unfortunately it is seldom utilized. The probable reasons for disregarding it are twofold: The mathematics involved appear overly complicated; and the quality levels suggested appear too loose. The procedure itself is based on a well established, if not well known, sampling test called sequential analysis. This paper describes the sequential analysis process. It examines the use of sequential analysis as a method of measuring the quality level of Test Program Sets. It describes how you can establish confidence levels on one will know what can be expected from fielded TPS's that have passed acceptance procedures.


    Access

    Access via TIB

    Check availability in my library


    Export, share and cite



    Title :

    Understanding and applying sequential analysis to TPS acceptance test procedures


    Additional title:

    Verstaendnis und Anwendung der sequentiellen Analyse bei der Akzeptanzpruefung von Pruefprogrammen


    Contributors:


    Publication date :

    1986


    Size :

    6 Seiten, 4 Bilder, 4 Tabellen, 7 Quellen


    Type of media :

    Conference paper


    Type of material :

    Print


    Language :

    English