Institute for Software Research International
School of Computer Science, Carnegie Mellon University


Modeling Techniques for a Risk Analysis Methodology
for Software Systems

Jim Wang

June 2003

Project Practicum
Master of Science in Information Technology in Software Engineering


Keywords: System specification, system modeling, UML, Z, EVES, Fault Trees

The U.S. Nuclear Regulatory Commission (NRC) Office of Nuclear Regulatory Research is interested in developing approaches towards analyzing digital instrumentation and control (I&C) systems for nuclear power plant system upgrades (Arndt 2002). These approaches are directed towards analyzing the changes in risk involved with using digital systems, which include software and hardware concerns. The purpose of this document is to outline possible techniques to use in a possible analysis methodology and to briefly describe how these techniques are used. This document suggests that a descriptive visual model be created using many of Unified Modeling Language's (UML) numerous artifacts. UML's widespread acceptance and comprehension, along with the possibility that UML documents have already been written for the system in question, make it an obvious choice for this application. In addition, a formal specification should be written in Z and analyzed with the help of the Z/EVES software package. Formal specification and analysis of that specification would ensure that the system is complete and that predetermined conditions hold throughout the operation of the software system. Finally, dynamic fault tree analysis should be performed on the system to analyze each hazard or failure event. Fault tree analysis is a technique that the NRC is very familiar with and adept at performing.

42 pages

Return to: SCS Technical Report Collection
School of Computer Science homepage

This page maintained by