|
|
Riskspectrum Analysis Tools (RSAT)
General
Features:
Analysis Tools (RSAT) is the analysis engine used by all of our products.
The MCS analysis is based on an improved version of the RSMCS algorithm.
The RSMCS algorithm is a top-down algorithm using features such as fault
tree structure optimization, modularisation and probabilistic cutoff. The
RSMCS algorithm has been widely recognized as the fastest and most
powerful existing fault tree solution algorithm.
RSMCS has been completely rewritten in the C language as a part of the
development of the WindowsTM
version of RiskSpectrum. Compared to the earlier version of RSMCS, the new
32-bit C version has the following main features:
- An improved code
structure, facilitating software maintenance and further development.
- Improved efficiency of
algorithm.
- Ability to handle event
tree structures. The application can handle individual sequence top
events and consequence top events.
In sequence top events, the application builds an internal sequence
fault tree structure, automatically linking all fault trees included in
the sequence, including the success logic.
In consequence top events, the program builds a large OR-gate with the
sequence fault trees for all sequences ending with the specified
consequence as input. The application can also automatically handle
linking between different event trees.
- Ability to handle
exchange events, i.e. replace basic events and/or gates triggered by
TRUE house events.
- Ability to handle
automated CCF modeling and quantification through CCF groups.
- Ability to handle NOTed
basic events and house events, NOR-gates and NAND-gates.
- Relative cutoff value in
addition to an absolute cutoff. The relative cutoff is dynamically
updated as a specified fraction of the sum of probabilities for
currently determined MCSs (during MCS generation).
- Automatic increase of
cutoff if "Too many MCS" is reached. This is repeated until the analysis
is completed.
- No absolute limits in
number of basic events, gates, MCS, etc.
Uncertainty Analysis:
The uncertainty analysis is based on Monte-Carlo simulation. The
simulation process is performed by simulating parameter uncertainty
distributions. The sampled parameter values are then used in
re-quantifying basic events and CCF events, and these are in turn used to
re-quantify the MCS equation.
Importance Analysis
Importance calculation can be performed for:
- Individual basic events
and CCF events
- CCF groups
- Parameters
- Attributes (groups of
basic events sharing the same attribute)
- Basic event groups
- Components
- Systems
- Initiating events
In each case, the following importance values can be calculated:
- Fussell-Vesely
- Fractional Contribution
- Risk Reduction
- Risk Achievement
Sensitivity Analysis
Sensitivity calculation can be performed for:
- Individual basic events
and CCF events
- CCF groups
- Parameters
- Attributes (groups of
basic events sharing the same attribute)
- Basic event groups
- Components
- Systems
- Initiating events
In each case, the following
sensitivity values can be calculated (sensitivity factor given by user):
- Sensitivity low (Top
probability, frequency …) when event divided with sensitivity factor
- Sensitivity high (Top
probability, frequency …) when event multiplied with sensitivity factor
- Sensitivity (Sens. High
/ Sens. Low)
Capabilities
Analysis Tools has no "hard" limit in number of gates, basic events,
minimal cut sets (MCS), etc. The total limit is defined by the available
RAM on the system running the analysis. For practical reasons, some of the
maximum limits will not be determined completely dynamically at run-time,
but will be read from a configuration file that can be controlled by the
user. For example, the user can control the upper limit in number of MCSs.
Back to top
|
