Data-Based Generation of Reliability Models
Aim: Design and implementation of methods for automated derivation of reliability models from data (e.g. A fault tree).
Objectives: Design a reliability model to produce streams of data. Analyze data for patterns using artificial intelligence and/or machine learning. Derive and reproduce the reliability model from the data. Compare the original reliability model and reproduced reliability model and evaluate the method of automated derivation.
The following demo showcases the method I developed to reconstruct a fault tree using time series data.
For a more in-depth explanation of the method please refer to my thesis which can be downloaded here
To look at the source code for the algorithms click here
Fault Tree Analysis
Original Fault Tree
Reconstructed Fault Tree
Minimal Cut Sets
- {{ set }}
Event
Metric
{{ formatEventName(chosenEvent)}}
| Fault Tree: | Original | Reconstructed |
|---|---|---|
| Reliability Distribution: | ||
| Maintainability Distribution: | ||
| Mean Time To Failures: | ||
| Mean Time To Repairs: | ||
| Operational Availability: |
