Application of Multivalued Test Sequencing to fault diagnosis

Abstract

In this paper, we consider a multivalued test sequencing (MVTS) problem, in which a test may have an arbitrary number of possible outcomes denoting different behaviors. The multivalued test sequencing problem can be solved through dynamic programming to find the optimal test sequence, but the time and space is large. In order to deal with this problem, we propose multivalued AO* algorithm, called MVAO*, based on multivalued heuristic evaluation function (MVHEF) in the AO*. With one example we demonstrate that MVAO* is efficient and effective for optimizing test sequences problem with multiple test responses.