pyRankMCDA

Introduction

pyRankMCDA is a Python library designed for rank aggregation in multi-criteria decision analysis (MCDA). It provides implementations of classical and modern rank aggregation methods, allowing users to combine multiple rankings into a single consensus ranking. This is particularly useful in fields like decision science, information retrieval, and any domain where synthesizing different ordered lists is necessary.

Citation

PEREIRA, V.; BASILIO, M.P.; FATIH Y. (2026). Unifying Multiple MCDA Rankings: Aggregation of Rankings through Methodological and Computational Perspectives. International Journal of Information Technology & Decision Making. doi: https://www.worldscientific.com/doi/10.1142/S0219622026500525

Features

• Multiple Rank Aggregation Methods:

  • Borda Method
  • Copeland Method
  • Footrule Rank Aggregation
  • Fast Footrule Rank Aggregation
  • Kemeny-Young Method
  • Fast Kemeny-Young
  • Median Rank Aggregation
  • PageRank Algorithm
  • Plackett-Luce Model Aggregation
  • Reciprocal Rank Fusion
  • Schulze Method

• Distance and Correlation Metrics:

  • Cayley Distance
  • Footrule Distance
  • Kendall Tau Distance
  • Kendall Tau Correlation
  • Spearman Rank Correlation

• Visualization Tools:

  • Heatmaps of rankings
  • Radar charts comparing rankings from different methods
  • Multidimensional Scaling (MDS) plots for visualizing distances between ranking methods

Usage

1. Install

pip install pyRankMCDA

2. Basic Example

import numpy as np
from pyRankMCDA.algorithm import rank_aggregation

# Example rankings from different methods

ranks = np.array([
    [1, 2, 3],
    [2, 1, 3],
    [3, 2, 1]
])

# Initialize rank aggregation object
ra = rank_aggregation(ranks)

# Run Borda method
borda_rank = ra.borda_method(verbose = True)

3. Running Multiple Methods

# Define the methods to run
methods = ['bd', 'cp', 'fky', 'md', 'pg']

# Run selected methods
df = ra.run_methods(methods)

4. Visualization

# Plot heatmap of rankings
ra.plot_ranks_heatmap(df)

# Plot radar chart of rankings
ra.plot_ranks_radar(df)

5. Computing Metrics

# Calculate distance and correlation metrics
d_matrix = ra.metrics(df)

# Plot metric comparisons
ra.metrics_plot(d_matrix)

6. Try it in Colab:

• Example: ( Colab Demo )

7. Others

• 3MOAHP - Inconsistency Reduction Technique for AHP and Fuzzy-AHP Methods
• pyDecision - A library for many MCDA methods
• pyMissingAHP - A Method to Infer AHP Missing Pairwise Comparisons
• ELECTRE-Tree - Algorithm to infer the ELECTRE Tri-B method parameters
• Ranking-Trees - Algorithm to infer the ELECTRE II, III, IV and PROMETHEE I, II, III, IV method parameters
• EC-PROMETHEE - A Committee Approach for Outranking Problems
• MCDM Scheduler - A MCDM approach for Scheduling Problems