ror (version 1.2)

ror-package: Robust Ordinal Regression MCDA sampler/solver

Description

This package implements UTAGMS and RORSMAA MCDA methods for ranking multiple alternatives in terms of multiple criteria. The current version assumes ascending preferences, i.e. higher criterion evaluation means higher preferability (=better).

References

Greco, S., Mousseau, V., Slowinski R., 2008. Ordinal regression revisited: multiple criteria ranking using a set of additive value functions. Eur J Oper Res 191 (2), 415.

Kadzinski, M., Tervonen, T., 2012. Stochastic Ordinal Regression for Multiple Criteria Decision Support. Submitted manuscript.

See Also

rorsmaa,utagms,sample.vfs.gibbs,sample.vfs.rejection

Examples

# Set Java VM memory use to 2g not to run out of heap space
options( java.parameters = "-Xmx2g" )
library(ror)

## Function needed to generate pareto-optimal alternatives
randomPointFromHypersphere <- function(ncrit) {
  rns <- c()
  while(TRUE) {
    rns <- rnorm(ncrit)
    if (all(rns > 0)) {
      break
    }
  }
  mul <- 1 / sqrt(sum(rns * rns))
  return(rns * mul)
}

performances <- t(replicate(10, randomPointFromHypersphere(5)))  # 10 alts, 5 crit
preferences <- matrix(c(1, 2, 4, 5, 7, 8, 1, 3), ncol=2, byrow=TRUE)

## Necessary relation
utagms(performances, preferences, necessary=TRUE, strictVF=TRUE)
## Possible relation
utagms(performances, preferences, necessary=FALSE, strictVF=TRUE)

## RORSMAA giving the POIs and RAIs
ror <- rorsmaa(performances, preferences)
print(ror$poi)
print(ror$rai)
cat(ror$misses, "misses while generating 10k value functions")

## Sample some value functions
vfs <- sample.vfs.gibbs(performances, preferences, nr=10, thinning=2)