Reliability demonstration testing for software
Hiroaki Sandoh · IEEE Transactions on Reliability · 1991
Reliability demonstration testing for software conducted to demonstrate whether the specified reliability has been achieved in a software product is studied. The software-reliability demonstration test is based on two risks. This procedure determines a test time and a maximum number of software failures during the test under prespecified producer and consumer risks. A procedure for zero software-failure reliability demonstration testing is proposed where the software product is accepted only if no failures occur in a specified testing time. The procedure based on two types of risk is identical to that for hardware items. The formulation for the zero software-failure reliability demonstration testing is identical to zero failure reliability demonstration testing for hardware items.>