Risk Research for Safety Critical Systems at the Technical University of Denmark

The talk will provide an overview of research projects that the risk research group of DTU has carried out, is working on and having in plans. The following topics and results will be briefly presented: a. Reliability models generalized to interval-valued probabilistic measures. In particular, reliability models with arbitrary connections of components, structural (stress-strength) reliability, statistical inference for reliability, and the use of the piece-wise exponential distribution family for reliability assessments. b. The use of discrete event simulation for the risk analysis of safety-critical systems and, in particular, the modelling of human performance for predicting failures and procedure violations. c. Capabilities-based approach for assessing the resilience of critical infrastructure. d. Identification of risks stemming from new communication technologies: 1. Hazard and operability analysis, independent of the system representation, 2. Combining formal correctness methods with hazard identification, 3. Formalization and automation of hazard and operability analysis for information systems. e. Unforeseen events with high impacts: validation of practices and models for predictability. f. Simulation based tool for risk assessment and mitigation in complex systems with strategic components.

Speaker Bio: Igor Kozine studied at the Moscow Institute of Physics and Engineering (Technical University) and received his M.S. (1983) and Ph.D. (1989) in Systems Sciences from the same university. Since then he worked at the Institute of Nuclear Power Engineering, Obninsk, Russia as associate professor and senior scientist. For two years (1994-1996) worked at Risø National Laboratory, Denmark, as a guest scientist, and for one year did a research study in the State University of New York at Binghamton as a Fulbright Scholar (1998-1999). From 1999 until now he has been working at the Technical University of Denmark as a Senior Researcher. His research is concerned with reliability, risk and uncertainty analysis as well as simulation of human performance. In particular, the use of imprecise statistical reasoning in relation to risk and reliability analysis has been the activity of his great interest for more than a decade. He has been applying the results of his research in very different domains: nuclear, wind power generation, oil and gas production and transport, maritime, railway systems and other critical infrastructure systems.

Presentation Slides: PDF

Note: Visitors from outside NIST must contact Cathy Graham; (301) 975-3800; at least 24 hours in advance.