Multicriteria Package

Provides a methodology to solve most of multicriteria ranking problems using partial and total pre-order from Promethee methods. Albuquerque & Montenegro (2015) .


RMCriteria is a package to solve Multiple-Criteria Decision Analysis (MCDA) problems. For now, it only supports Promethee methods, but other methods may be developed in the future.


Currently, the easiest way to install RMCriteria is through devtools package, straight from GitHub:

# install.packages("devtools")


Using RMCriteria is quite simple. The general idea is to create a RPrometheeArguments object, with all parameters such as the criterias and alternatives and then applying this object to the chosen method, like RPrometheeI.

## Create objects for each argument
data <-matrix(c(5.2,-3.5,
                6.7,-2.0), byrow = T, ncol=2, nrow=3)
parms <- matrix(c(NA,
vecWeights <- c(0.3,0.7)
vecMaximiz <- c(F,T)
prefFunction <- c(0,0)
normalize <- FALSE
alternatives <- c("Alt 1", "Alt 2", "Alt 3")
## Create RPrometheeArguments object
PromObj <- RPrometheeConstructor(datMat = data, vecWeights = vecWeights, vecMaximiz = vecMaximiz, prefFunction = prefFunction, parms = parms, normalize = normalize, alternatives = alternatives)
# Run RPrometheeI
(result <- RPrometheeI(PromObj))

More Information

RMCriteria was developed in the Laboratory of Machine Learning in Finance and Organizations (LAMFO) from University of Brasilia, in Brazil. LAMFO is a center devoted to research machine learning methods and related subjects applied to organizations in Marketing, Finance, Logistics and many others.


Reference manual

It appears you don't have a PDF plugin for this browser. You can click here to download the reference manual.


0.2.0 by Pedro Albuquerque, 2 months ago

Report a bug at

Browse source code at

Authors: Pedro Albuquerque and Gustavo Monteiro

Documentation:   PDF Manual  

GPL-2 license

Imports Rcpp, RcppNumerical, linprog, lpSolve, ggplot2, gridExtra, ggnetwork, network, pastecs, methods, graphics, stats, utils, dplyr

Linking to Rcpp, RcppEigen, RcppNumerical

System requirements: GNU make

See at CRAN