Multiple-criteria-decision-analysis

The aim of this project is to propose a turnkey implementation of the main multi-criteria decision-support tools.

This project was created by Marin CHEVOLLEAU helped by Tristan MANIER.

Requirements

Python dependencies

• pandas
• typing
• networkx
• matplotlib

Structure

Files

├─ 📒 data → input datasets (example of cities to rank)
├─ 📒 output → csv result files
├─ 📜 README.md → This file
├─ 🐍 criterion.py → criterion class
├─ 🐍 normalize.py → normalize function
├─ 🐍 dominance.py → find the dominant pareto solutions
├─ 🐍 satisfaction.py → find satisfying solutions according to the decision-maker needs
├─ 🐍 electre_1.py → plot ranking graph using Electre 1 method
├─ 🐍 topsis.py → rank solutions using TOPSIS method
└─ 🐍 weighting.py → rank solutions using weighted sum method

Execution

Global

Run the main script main.py. Uncomment lines to select the satisfaction or dominance method to use.

Individual

First you need to run python3 dominance.py OR python3 satisfaction.py

Then you can use the following (independant) scripts:

• python3 weighting.py
• python3 electre_1.py
• python3 electre_2.py
• python3 topsis.py