Getting started
PySensMCDA
PySensMCDA
is a comprehensive Python package tailored specifically for Multi-Criteria Decision Analysis (MCDA) sensitivity analysis. MCDA is a powerful tool used in decision-making processes to evaluate alternatives based on multiple conflicting criteria. PySensMCDA
empowers users to delve deeper into the robustness and reliability of their decision models by exploring the sensitivity of results to variations in input parameters.
In essence, this package offers tools for:
Decision matrix sensitivity analysis
Weights sensitivity analysis
Ranking sensitivity analysis
Perturbation generation
Weights generation
Visualizations of sensitivity analysis
Installation
The package can be download using pip:
pip install pysensmcda
Testing
The modules performance can be verified with pytest
library
pip install pytest
pytest tests
Modules and functionalities
Alternative:
Name |
Reference |
---|---|
Discrete modification |
- |
Percentage modification |
|
Range modification |
- |
Alternative removal |
Criteria:
Name |
Reference |
---|---|
Random distribution - weights generation |
- |
Chisquare distribution |
- |
Laplace distribution |
- |
Normal distribution |
- |
Random distribution |
- |
Triangular distribution |
- |
Uniform distribution |
- |
Percentage modification |
|
Range modification |
- |
Weights scenarios |
- |
Criteria identification |
|
Criteria removal |
Probabilistic:
Name |
Reference |
---|---|
Monte carlo weights generation |
|
Perturbed matrix |
|
Perturbed weights |
Ranking:
Name |
Reference |
---|---|
Ranking alteration |
|
Demotion |
- |
Promotion |
|
Fuzzy ranking |
- |
Compromise:
Name |
Reference |
---|---|
Borda |
|
Improved Borda |
|
Dominance directed graph |
|
Half-quadratic compromise |
|
ICRA - Iterative Compromise Ranking Analysis |
|
Rank position method |
Graphs:
Name |
---|
Heatmap |
Promotion-demotion ranking graph |
Preference distribution |
Rankings distribution |
Values distribution |
Weights barplot |
Usage example
General usage examples see
examples.ipynb
Graphs submodule usage examples see
graphs_examples.ipynb
Literature example analysis see
literature_example.ipynb