RELIABILITY ESTIMATION FROM APPROPRIATE TESTING OF PLANT-PROTECTION SOFTWARE

Plant protection software may be realistically tested using inputs fro m a plant model before its initial use, or when it is not feasible to take the plant into certain fault conditions. If statistical estimatio n of software reliability is to be performed using the test results, i t is not sufficient for the plant model to produce inputs which are si mply correct in the sense that the plant could have produced them. In addition, the operational distribution of the input space must be simu lated. The paper illustrates how to perform such a simulation, by deve loping an example in which an existing non-random plant model is rando mised to simulate the operational distribution of the software. In add ition, two methods of estimating the probability of failure on demand (pfd) for a program are reported. Both methods estimate a pfd given re sults from dynamic testing, during which the program is exercised acco rding to its operational distribution. The first method is standard an d has been used previously in the context of software testing. The sec ond estimation method has been developed recently within a program of Nuclear Electric research into software reliability testing. The disti nguishing foundational assumptions of the two methods are discussed.