ML-CB-Bipolar-L-Fuzzy-Rough-PROMETHEE

This repository contains the implementation of a machine learning-enhanced PROMETHEE model using a Covering-Based Bipolar L-Fuzzy Rough Set approach for evaluating battery energy storage systems (BESS) in renewable energy projects.

🔬 This code supports the case study presented in the published article:
"An Enhanced Machine Learning Covering-Based Bipolar L-Fuzzy Rough Set PROMETHEE Model for Battery Storage Systems in Renewable Energy", published in Expert Systems with Applications (Elsevier).
DOI: 10.1016/j.eswa.2025.127951

---

✨ Highlights

• 📊 Uses Random Forest to compute criteria weights from data
• 🔁 Applies Covering-Based Bipolar L-Fuzzy Rough Sets for modeling uncertainty
• ⚖️ Uses PROMETHEE for preference ranking and decision-making
• 🔌 Real-world application for evaluating battery storage systems in renewable energy

---

📁 Files

• ml_promethee_decision_model.ipynb: Main Jupyter notebook with full workflow
• requirements.txt: Python libraries needed to run the notebook
• graph.png: Graphical figure used in the paper (optional)
• output.gif: Animated decision visualization (optional)

---

🧠 Methodology Workflow

1. Random Forest is trained on input data to determine feature importance (criteria weights)
2. Covering-Based Bipolar L-Fuzzy Rough Sets are used to model uncertainty and derive approximations
3. PROMETHEE is applied to generate preference flows and rankings for decision alternatives

---

🚀 How to Run

1. Clone the repository:

   git clone https://github.com/Faiza-Tufail/ML-CB-Bipolar-L-Fuzzy-Rough-PROMETHEE.git
   cd ML-CB-Bipolar-L-Fuzzy-Rough-PROMETHEE
   

2. Install required libraries:

   pip install -r requirements.txt

3. Launch the Jupyter notebook:

   jupyter notebook ml_promethee_decision_model.ipynb

4. Follow the notebook cells to run the analysis and reproduce the results.

---

📚 References

• Faiza Tufail et al., An Enhanced Machine Learning Covering-Based Bipolar L-Fuzzy Rough Set PROMETHEE Model for Battery Storage Systems in Renewable Energy, Expert Systems with Applications, 2025.
  DOI: 10.1016/j.eswa.2025.127951

---

🔋 Normalized Performance of Different Batteries

The following graph shows the normalized performance of various battery storage systems based on selected evaluation criteria.

---

🏆 Final Ranking for Best Battery (Random Forest)

The animation below demonstrates how the Random Forest model contributes to the final ranking of the most suitable battery for renewable energy systems.

📝 License

This repository is licensed under the MIT License.

---

🙋 Contact

For questions or feedback, please contact Faiza Tufail at faizatufail85@gmail.com.