Test Case Generation as an AI Planning Problem

  • Authors:
  • Adele E. Howe;Anneliese von Mayrhauser;Richard T. Mraz

  • Affiliations:
  • Computer Science Department, Colorado State University, Fort Collins, CO 80523/ E-mail: howe&commat/cs.colostate.edu, E-mail: avm&commat/cs.colostate.edu;Computer Science Department, Colorado State University, Fort Collins, CO 80523/ E-mail: howe&commat/cs.colostate.edu, E-mail: avm&commat/cs.colostate.edu;HQ USAFA/DFCS, 2354 Fairchild Hall, Suite 6K41, U.S. Air Force Academy, CO 80840/ E-mail: mraz&commat/cs.usafa.af.mil

  • Venue:
  • Automated Software Engineering
  • Year:
  • 1997

Quantified Score

Hi-index 0.00

Visualization

Abstract

While Artificial Intelligence techniques have been appliedto a variety of software engineering applications, the area ofautomated software testing remains largely unexplored. Yet, testcases for certain types of systems (e.g., those with command languageinterfaces and transaction based systems) are similar to plans. We have exploitedthis similarity by constructing an automated test case generator withan AI planning system at its core. We compared the functionality andoutput of two systems, one based on Software Engineering techniquesand the other on planning, for a real application: the StorageTekrobot tape library command language. From this, we showed that AIplanning is a viable technique for test case generation and that thetwo approaches are complementary in their capabilities.