Software Test Process
Original Publication Date: 1989-Jan-01
Included in the Prior Art Database: 2005-Jan-27
A system often has so many hardware attachment possibilities that it is impossible to "verify" software in all possible combinations of attachments. A process is described in which a "reasonable" number of rigorously selected attachment combinations are evaluated to determine what main effects can cause significant problems. The intent of this "initial" experiment is not to consider interaction effects or even repeat main effect levels many times when these main effects have a large number of levels. However, this process does yield a structured matrix of variable combinations that should capture problems quicker than the classical "one at a time" approach.