Complete functional testing

Mike Holcombe

(University of Sheffield)

All software systems are subject to testing - for some of them, particularly safety-critical systems, testing is the major activity in the project. The principal purpose of testing is to detect faults in a software system (to enable their removal). However, no existing methods allow us to make any statement about the type or precise number of faults that remain undetected after testing is completed. In particular we are unable to state that specific components of the system are free from fault after testing has been concluded. We demonstrate that a new method for generating test cases allows us to make sensible claims about the level and type of faults remaining after the testing process is complete. The method is illustrated by a case study.
Tuesday 23rd January 1996, 14:30
Seminar Room 322
Department of Computer Science