Expert System for Gaza Post-War Building Assessment

🌐 Access the Deployed App

This project implements a rule-based expert system to assist engineers in assessing post-war building conditions in Gaza. The system evaluates various structural, environmental, social, and utility-related factors to prioritize actions for reconstruction, repair, or temporary solutions.

---

Features

• Rule-Based Inference: Implements 40+ rules to assess building conditions based on predefined criteria.
• Uncertainty Handling: Uses fuzzy logic, confidence levels, and priority ranking for accurate recommendations.
• Prioritization Mechanism: Outputs are ranked by severity and relevance.
• Professional UI: A polished Streamlit-based web interface with logical grouping, dynamic inputs, and integrated social links.
• Robust Testing: Includes tests for individual rules, combined scenarios, and edge cases, ensuring rule validation.
• Successful Deployment: Fully deployed and accessible online.

---

Project Structure

• src/: Contains the main expert system implementation.
  • building_assessment_ES.py: Implements rule-based logic for building assessment.
  • building_assessment_UI.py: Streamlit-based user interface for user interaction and result visualization.
• test/: Includes test cases and a validation notebook.
  • testing.ipynb: Jupyter Notebook for individual and combined rule testing.
• docs/: Project documentation and supporting files.
  • Final Report - A Rule-Based Expert System for Post-War Building Assessment in Gaza (2025).pdf: Final report detailing the system and its development.
  • Presentation - Rule-Based Expert System for Post-War Building Assessment in Gaza: PowerPoint presentation summarizing the project.

---

Dependencies

This project requires the following dependencies:

• Streamlit: For building the user interface.
• NumPy: Numerical computations.
• SciPy: Used by scikit-fuzzy for advanced computations.
• Scikit-Fuzzy: For fuzzy logic implementation.
• Custom Fork of Experta: A modified version of the experta library compatible with Python 3.10+. Install via:

  git+https://github.com/WalidAlsafadi/experta.git
  

Setup Instructions

1. Clone the repository:

   git clone https://github.com/WalidAlsafadi/BuildingAssessment-ExpertSystem

2. Navigate to the project directory:

   cd BuildingAssessment-ExpertSystem

3. Install the required dependencies:

   pip install -r requirements.txt

Usage

Running the Expert System via the UI

1. Launch the Streamlit UI:

   streamlit run src/building_assessment_UI.py

2. Follow the interactive interface to input building conditions and view prioritized actions.

Running the Expert System via Python

1. Import the system into your script:

   from src.building_assessment_ES import BuildingAssessment, BuildingAssessmentExpertSystem

2. Define input facts and run the engine:

   engine = BuildingAssessmentExpertSystem()
   engine.reset()
   engine.declare(BuildingAssessment(hazardous_zone=True, overcrowding=True))
   engine.run()
   engine.print_prioritized_actions()

Testing

1. Open the Jupyter Notebook in the test folder:

   jupyter notebook test/testing.ipynb

2. Run individual and combined scenario tests.
3. Validate the outputs in the notebook.

How It Works

• The system uses rule-based inference with 40+ predefined rules to assess building conditions.
• Users input building conditions through a user-friendly UI.
• The system processes these inputs and provides actionable recommendations based on priority and confidence.

Future Enhancements and Updates

As this expert system evolves, future improvements may include:

• Expanded Rule Base: Adding new rules to address emerging challenges in post-conflict zones.
• Machine Learning Integration: Predict building damage or prioritize reconstruction based on historical data.
• Enhanced User Experience: Improving the UI for better usability and multi-language support.
• Real-Time Data: Integration of real-time data (e.g., satellite imagery) for more precise assessments.

Authors

• Walid Alsafadi

License

This project is licensed under the Apache License. See LICENSE for details.

---

Let me know if further refinements are needed! 😊