Derivation of a Suitable Finite Test Suite for Customized Probabilistic Systems

In order to check the conformance of an IUT (implementation under test) with respect to a specification, it is not feasible, in general, to test the whole set of IUT available behaviors. In some situations, testing the behavior of the IUT assuming that it is stimulated by a given usage model is more appropriate. Specifically, if we consider that specifications and usage models are defined in probabilistic terms, then by applying a finite set of tests to the IUT we can compute a relevant metric: An upper bound of the probability that a user following the usage model finds an error in the IUT. We also present a method to find an optimal (with respect to the number of inputs) set of tests that minimizes that upper bound.