Decerns MCDA is a multi-criteria decision analysis system, developed as a standalone/desktop application (Desktop Edition). The system implements a variety of methods/models for multi-criteria decision analysis (MCDA) along with the tools for multi-criteria problems structuring and modeling (Value Tree and Performance Table), weighting criteria, sensitivity analysis and uncertainty treatment.

When solving a specific multi-criteria problem, a DECERNS user has the opportunity to choose the appropriate MCDA method(s). If desired, the user may compare results after the implementation of several MCDA methods.

The following multi-criteria methods/models and tools has been implemented in the Decerns MCDA:

basic methods such as

  • MAVT (Multi-Attribute Value Theory);
  • AHP (Analytic Hierarchy Process);
  • TOPSIS (Technique for Order Preference by Similarity to the Ideal Solution); and
  • PROMETHEE (Preference Ranking Organization METHod for Enrichment Evaluations)

advanced methods such as

  • MAUT (Multi-Attribute Utility Theory);
  • FlowSort (for analysis of sorting multicriteria problems);
  • ProMAA (Probabilistic Multi-criteria Acceptability Analysis)

and extensions of methods based on fuzzy set approaches

  • F-MAVT (Fuzzy MAVT), and
  • FMAA (Fuzzy MAA)

The following tools are also used within the MCDA-subsystem:

  • Value Tree and Performance Table for problem structuring (and data input and presentation);
  • Value Path for source data presentation;
  • Scatter Plot for presentation of source data in 2-dimensional criteria space;
  • Weight Sensitivity analysis (line weights and walking weights tools);
  • Value function sensitivity analysis (for MAVT, MAUT, ProMAA, F-MAVT, FMAA methods);
  • Different graphical and tabular tools for probabilistic input/output data presentation (probability and density distributions), including setting source probability distributions for criteria (MAUT, ProMAA) and weights (ProMAA) with the use of Normal, Uniform, Log-Normal and delta distributions;
  • Different graphical and tabular tools for fuzzy input/output data presentation, including setting source fuzzy numbers (triangular, trapezoidal, piecewise linear, singleton) for criteria and weights (F-MAVT, FMAA).

Uncertainties are addressed through sensitivity analysis for weight coefficients, and using random performance (for MAUT), random weights and random performance (ProMAA), and, accordingly, fuzzy weights and fuzzy performance (for F-MAVT and FMAA).

Technologies: The system is implemented as a desktop application in Java (requires an installed Java Runtime Environment). The application uses interactive components based on Java2D and special libraries jSimpleGraph and jSimpleViz.