ELECTRE Tri-B

by Christian Ghiaus

Researcher ID: ORCID, SciProfiles, Scopus, Web of Science, HAL

ELECTRE Tri-B is a Multiple-Criteria Decision-Making (MCDM) method for sorting alternatives into predefined categories defined by lower and upper boundaries.

This repository contains a framework and a Python 3.11.7 implementation of ELECTRE Tri-B and of Simos' method for determining the weights of criteria, along with documentation and examples.

The crisp and probabilistic ELECTRE Tri-B methods are implemented in the module electre_tri.py. Four workflows are presented in workflows.ipynb for crisp and probabilistic ELECTRE Tri-B methods using base profiles defined explicitly or by their worst and best levels.

Installation

This software requires no installation. Simply copy the module files to your desired location.

For remote execution without any local setup, you can run this repository on MyBinder.org:

Click the "Launch Binder" badge above to open the repository in a Jupyter environment. This allows you to interact with the software directly in your web browser (with no local installation).

Contents

ELECTRE_Tri-main
├── code
│   └── examples
├── data
├── docs
│   ├── explanation
│   ├── how_to_guides
│   ├── reference
│   └── tutorials
├── figs
├── src
│   ├── __init__.py
│   ├── electre_tri.py
│   └── simos_revised.py
├── LICENSE
├── README.md
├── environment.yml
└── requirements.txt

Summary

• code/examples: Python scripts that use the modules from /src and data from /data.

• data: .csv data files for ELECTRE Tri-B and Simos' revised method.

• docs: documentation

  • explanation: description of concepts; details on ELECTRE Tri-B and Simos' method for weights of criteria.
  • how_to_guides: directions for specific goals; useful to solve tasks when accustomed with the methods.
  • reference: documentation of functions implemented in the modules of /src.
  • tutorials: guided steps to solve a meaningful problem; starting point for those new to the methods.

• figs: modifiable figures in .svg format.

• src: source code of Python modules.

  • electre-tri.py: Multiple-Criteria Decision-Making.
  • simos_revised: Simos' method for determining weights of criteria in ELECTRE type methods.

Usage

1. Launch binder.
2. Upload your data file (which describes the ELECTRE Tri or the weighing problem) in /data folder.
3. Modify and run the scripts from code/examples to get the results.

Read the documentation from /docs.