Despite improvements in software engineering processes and tools, concrete preventative and analytical software quality assurance activities are still typically manually triggered and determined, resulting in missed or untimely quality opportunities and increased project overhead. While some processes specify abstract quality assurance measures, concrete requisite measures directly relevant for specific product artifacts (e.g., code) or processes (e.g., testing) must be determined operationally and contemporaneously, yet are hitherto often determined manually and unsystematically. Quality goals, when defined, lack holistic environmental support for automated attainment measurement and governance that is tightly integrated in the low-level operational software engineering processes, resulting in higher quality risks and cost risks. This presentation gives insight into efforts taken in the Q-ADVICE project to improve that situation. Based on adaptive process management, an approach is presented that tightly integrates situationally-determined quality measure proposals into the concrete software developer workflow, using contextual semantic knowledge and multi-agent quality goal tracking and decision making. The evaluation shows the feasibility of the approach for automatically providing timely quality measure guidance to software engineers without disrupting their current activity. Such an approach can support process governance while reducing quality risks and costs during software development projects.
DBIS-Kolloquium, Gregor Grambow, Ort: O27/545, Zeit: 15:00 Uhr, Datum: 06. April 2011