Risk and Reliability Analysis: FTA, ETA, and Bayesian Networks

A collection of Jupyter notebooks demonstrating three classical risk and reliability analysis techniques applied to industrial, cyber-physical, and safety-critical case studies:

• Fault Tree Analysis (FTA) — top-down, deductive failure analysis
• Event Tree Analysis (ETA) — forward-looking accident progression analysis
• Bayesian Networks (BN) — probabilistic graphical models for causal reasoning under uncertainty

Each technique is presented at multiple complexity levels (simple, medium, complex) and applied to real-world inspired case studies including chemical plants, smart hospitals, autonomous vehicles, railway signalling, water treatment plants, infusion pumps, oil pipelines, and smart power grids.

Repository Structure

.
├── README.md
├── requirements.txt
│
├── FTA_Simple_Autonomous_Vehicle.ipynb         # FTA - Autonomous Vehicle Emergency Stop Failure
├── FTA_mediumlevel_case_study.ipynb            # FTA - Smart Infusion Pump (medium)
├── FTA_Chemical plant_casestudy.ipynb          # FTA - Chemical Plant (complex)
├── FTA_Library_Autonomous_Vehicle_.ipynb       # FTA using pfta library - AV stopping failure
├── FTA_Library_Chemical_Plant_.ipynb           # FTA using pfta library - Chemical plant cyber-safety
├── FTA_library_Smart_Infusion_Pump.ipynb       # FTA using pfta library - Smart infusion pump
│
├── ETA_Railway_Signal_Failure.ipynb            # ETA - Railway Signal Failure at a Junction
├── ETA_mediumlevel.ipynb                       # ETA - Cyber-Induced Valve Control Failure (Oil Pipeline)
├── ETA_complexlevel.ipynb                      # ETA - Cyber-Induced Failure in Smart Power Grid Substation
│
├── BN_Smart_Hospital.ipynb                     # BN - Smart Hospital Monitoring System
├── BN_Smart_Water_Treatment.ipynb              # BN - Smart Water Treatment Plant Cyber-Safety
└── BNComplexlevel.ipynb                        # BN - Cyber-Physical Risk in Chemical Processing Plant

Notebook Summaries

Fault Tree Analysis (FTA)

Notebook	Case Study	Level
FTA_Simple_Autonomous_Vehicle.ipynb	Autonomous vehicle emergency-stop failure	Simple
FTA_mediumlevel_case_study.ipynb	Smart infusion pump failure	Medium
FTA_Chemical plant_casestudy.ipynb	Chemical plant top-event failure	Complex
FTA_Library_Autonomous_Vehicle_.ipynb	Autonomous vehicle stopping failure (uses pfta library)	Library-based
FTA_Library_Chemical_Plant_.ipynb	Chemical plant cybersecurity-safety failure (uses pfta)	Library-based
FTA_library_Smart_Infusion_Pump.ipynb	Smart infusion pump cybersecurity-safety failure (uses pfta)	Library-based

Event Tree Analysis (ETA)

Notebook	Case Study	Level
ETA_Railway_Signal_Failure.ipynb	Railway signal failure at a junction	Simple
ETA_mediumlevel.ipynb	Cyber-induced valve control failure in an oil pipeline pumping station	Medium
ETA_complexlevel.ipynb	Cyber-induced failure in a smart power grid substation (6 defensive layers)	Complex

Bayesian Networks (BN)

Notebook	Case Study	Level
BN_Smart_Hospital.ipynb	Smart hospital monitoring failure leading to patient risk	Simple
BN_Smart_Water_Treatment.ipynb	Cyber-safety risk in a smart water treatment plant	Medium
BNComplexlevel.ipynb	Cyber-physical risk in a chemical processing plant (8 nodes, 10 edges)	Complex

Each notebook includes the model construction, probability calculations, visual diagrams (event trees, fault trees, or DAGs), inference queries, summary tables, and an interpretation of the results.

Requirements

• Python 3.9 or newer
• Jupyter Notebook or JupyterLab (or Google Colab)
• The system-level Graphviz binary (required by the Python graphviz wrapper used to render fault trees and event trees)

Install Graphviz on your system:

• Windows: Download the installer from graphviz.org/download and add the bin folder to your PATH
• macOS: brew install graphviz
• Ubuntu / Debian: sudo apt-get install graphviz

Then install the Python dependencies:

pip install -r requirements.txt

Running the Notebooks

Clone the repository and launch Jupyter:

git clone https://github.com/<your-username>/<your-repo-name>.git
cd <your-repo-name>
pip install -r requirements.txt
jupyter notebook

Then open any of the .ipynb files and run the cells top to bottom.

You can also run any notebook directly in Google Colab — open the notebook on GitHub and click "Open in Colab", or upload the file at colab.research.google.com.

Libraries Used

• pandas — result tables and data handling
• matplotlib — charts, icons, and plots
• networkx — graph construction and visualization
• pgmpy — Bayesian Network modelling and inference (DiscreteBayesianNetwork, TabularCPD, VariableElimination)
• graphviz — Python wrapper for rendering fault tree and event tree diagrams
• pfta — fault tree analysis library used in the FTA library-based notebooks
• ipython — for displaying images and rich output inside notebooks

Notes

• The notebooks were originally developed and tested in Google Colab. When running locally, ensure the Graphviz binary is installed and accessible on your PATH, otherwise diagrams will not render.
• Some notebooks include !pip install ... cells — these are safe to run in Colab and harmless locally if dependencies are already installed via requirements.txt.

Author

Aromal Email: aromalmanu03@gmail.com

License

This project is released under the MIT License. Feel free to use, modify, and share with attribution.