Defensio dissertationis, 20.12.2011
Knowledge Based Inspection: Modelling Complex Processes with the integrated Safeguards Medelling Method (iSMM)
Florin Abazi (Doktor der technischen Wissenschaften)
Öffentliche Verteidigung der Dissertation
20.12.2011, 10:00 Uhr
Increased level of complexity in almost every discipline and operation today raises the demand for knowledge in order to successfully run an organization whether to generate profit or to attain a non-profit mission. Traditional way of transferring knowledge to information systems rich in data structures and complex algorithms continue to hinder the ability to swiftly turnover concepts into operations. Diagrammatic modelling commonly applied in engineering in order to represent concepts or reality remains to be an excellent way of converging knowledge from domain experts.
The nuclear verification domain represents ever more a matter which has great importance to the World safety and security. Demand for knowledge about nuclear processes and verification activities used to offset potential misuse of nuclear technology will intensify with the growth of the subject technology. This Doctoral thesis contributes with a model-based approach for representing complex process such as nuclear inspections. The work presented contributes to other domains characterized with knowledge intensive and complex processes.
Based on characteristics of a complex process a conceptual framework was established as the theoretical basis for creating a number of modelling languages to represent the domain. The integrated Safeguards Modelling Method (iSMM) is formalized through an integrated meta-model. The diagrammatic modelling languages represent the verification domain and relevant nuclear verification aspects. Such a meta-model conceptualizes the relation between practices of process management, knowledge management and domain specific verification principles. This fusion is considered as necessary in order to create quality processes.
The study also extends the formalization achieved through a meta-model by
contributing with a formalization language based on Pattern Theory. Through
the use of graphical and mathematical constructs of the theory, process structures are formalized enhancing the ability to analyse, compare and transform models. In the example domain all possible connections between critical nuclear processes were formalized providing also for probability-based analysis of weapons acquisition paths that will help design objective-based inspection processes.