A history-based approach for early faulty state detection

Abstract

This paper presents a history-based approach to detect class faults that a test set may fail to uncover. The method sequence test history contains the run-time information about object states, state transitions, and dependence relations between data members. The approach examines the test history to determine if some faulty state (or new state) is reachable from the states that have been covered by the test set. The main benefit of the approach is that it only depends on the existing test set that has been executed, and hence it does not require the development of any new test case.