Industrial application of criticality predictions in software development

Cost-effective software project management has the serious need to focus resources on areas where they have biggest impact. Criticality predictions are typically applied for finding out such high-impact areas in terms of most effective defect detection. The paper investigates two related questions in the context of real projects, namely the selection of the best classification technique and the use of its results in directing management decisions. Results from a current large-scale switching project are included to show practical benefits. Fuzzy classification yielded best results and is in a second study enhanced with genetic algorithms to improve overall prediction effectiveness