objective-weights-for-mcda

This is Python 3 library dedicated for multi-criteria decision analysis with criteria weights determined by objective weighting methods. The documentation is provided here

Installation

Downloading and installation of objective-weights-mcda package can be done with using pip

pip install objective-weights-mcda

Methods

mcda_methods includes:

• vikor with VIKOR method

Other modules include:

• additions include rank_preference method for ranking alternatives according to MCDA score

• correlations include:

  • Spearman rank correlation coefficient spearman,
  • Weighted Spearman rank correlation coefficient weighted_spearman,
  • Pearson correlation coefficient pearson_coeff

• normalizations with methods for decision matrix normalization:

  • linear_normalization - Linear normalization,
  • minmax_normalization - Minimum- Maximum normalization,
  • max_normalization - Maximum normalization,
  • sum_normalization - Sum normalization,
  • vector_normalization - Vector normalization

• weighting_methods include 11 objective weighting methods for determination of criteria weights (significance) without decision-maker involvement:

  • equal_weighting - Equal weighting method
  • entropy_weighting - Entropy weighting method
  • std_weighting - Standard deviation weighting method
  • critic_weighting - CRITIC weighting method
  • gini_weighting - Gini coefficient-based weighting method
  • merec_weighting - MEREC weighting method
  • stat_var_weighting - Statistical variance weighting method
  • cilos_weighting - CILOS weighting method
  • idocriw_weighting - IDOCRIW weighting method
  • angle_weighting - Angle weighting method
  • coeff_var_weighting - Coefficient of variation weighting method

Examples of usage of objective_weights_mcda are provided on GitHub in examples

License

This package called objective-weights-mcda was created by Aleksandra Bączkiewicz. It is licensed under the terms of the MIT license.

Note

This project is under active development.