Tid: 12 mars 2012 kl 15.15-16.00.Seminarierummet 3721, Institutionen för matematik, KTH, Lindstedts väg 25, plan 7. Karta!
Föredragshållare: Pontus Johnson, Department of Industrial Information and Control Systems, KTH
Titel: Predictive, Probabilistic Architecture Modeling
Abstract: In the design phase of software system development, it is desirable to predict the properties of the system-to-be. Formalisms to assess such qualities as performance, reliability and security have therefore previously been proposed. Existing prediction systems do, however, not allow the modeler to express uncertainty with respect to the design of the considered system. As systems grow larger, older and more complex, our knowledge about them grows hazier. In many real system development projects today, uncertainty regarding the systems' present or future characteristics is so significant that ignoring it becomes problematic. In this talk, we propose a formalism, the Predictive, Probabilistic Architecture Modeling Framework (P2AMF), capable of advanced and probabilistically sound reasoning about system models given in the form of UML class and object diagrams. The proposed formalism is based on the Object Constraint Language (OCL). To OCL, P2AMF adds a probabilistic inference mechanism. The talk introduces P2AMF, describes its use for system property prediction and assessment, and proposes an algorithm for probabilistic inference. This is followed by the presentation of a software tool implementing P2AMF (publicly available for download), and by a comparison of P2AMF to related work.
|Sidansvarig: Filip Lindskog