Constraints: The Future of Combinatorial Interaction Testing

Abstract

Combinatorial Interaction Testing (CIT) has gained a lot of attention in the area of software engineering in the last few years. CIT problems have their roots in combinatorics. Mathematicians have been concerned with the NP-complete problem of finding minimal covering arrays (in other words, minimal CIT test suites) since early nineties. With the adoption of these techniques into the area of software testing, an important gap has been identified - namely consideration of real-world constraints. We show that indeed finding an efficient way of handling constraints during search is the key factor in wider applicability of CIT techniques.