Proceedings of the Eighth International Conference on Probabilistic Safety Assessment & Management (PSAM)
156 QRAS Approach to Phased Mission Analysis (PSAM-0444)
Download citation file:
- Ris (Zotero)
- Reference Manager
The Quantitative Risk Assessment System (QRAS) is a comprehensive PC-based tool for probabilistic risk assessment (PRA) of safety critical systems, commercialized from a NASA based software program. QRAS capabilities for conventional risk analysis of large and complex systems have been demonstrated in applications to Space Shuttle PRA and test trials with the International Space Station risk models. By design the system is also well suited to other systems such as ground vehicles, medical devices, and nuclear plants.
This paper presents a new and unique feature of QRAS, under implementation, to handle time-dependent occurrences of events across the time-phases of a system mission. Complex system's risk models are often divided into time-phases because of the way system functions or for ease of modeling. However, while analyzing the accident scenario in these time-phases it is necessary to model the component that function across the phases correctly. A usual assumption that all components are available at the start of every new phase can be overly optimistic and might give results that are significantly off the mark. QRAS would allow the user to model the time-phases, operational time interval information for each accident scenario and that in combination with start time and duration of basic events provides for a platform to incorporate probabilities of components having failed or degraded even before entering the subsequent phases.
Many techniques for handling phased-mission fault trees have been published before but their complexities have prohibited their use in mainstream PRA applications. QRAS has a platform to capture some of those techniques without substantially increasing the complexity of the solution such that a complete PRA analysis of a large complex system is still solvable, with its Binary Decision Diagrams (BDD) engine, with acceptable computational speed.