VIKOR {MCDA}

VIKOR method

Description

VIKOR is a multi-criteria decision analysis method originally developed by Serafim Opricovic in his 1979 Ph.D. Thesis, and later published in 1998.

Usage

VIKOR(
  performanceTable,
  criteriaWeights,
  criteriaMinMax,
  v = 0.5,
  positiveIdealSolutions = NULL,
  negativeIdealSolutions = NULL,
  alternativesIDs = NULL,
  criteriaIDs = NULL
)

Arguments

performanceTable

Information matrix with nAlt rows and nCrit columns. Values correspond to the level the corresponding criteria takes for the corresponding alternative. All values should be numeric. Rows and columns should be named as the alternatives and criteria, respectively.

criteriaWeights

Numeric vector with nCrit elements. Should be named.

criteriaMinMax

Character vector with nCrit elements. It should contain values "min" if the corresponding criteria is to be minimised (less is better), or "max" if the corresponding criteria is to be maximised (more is better).

v

Numeric scalar. Parameter defining the importance given to the group utility, with respect to the minimun regret of the opponent alternative. Should be between 0 and 1. Default is 0.5.

positiveIdealSolutions

Numeric vector of ideal criteria values. If omitted, then they are defined as the best values observed among the existing alternatives.

negativeIdealSolutions

Numeric vector of worst possible criteria values. If omitted, then they are defined as the worst values observed among the existing alternatives.

alternativesIDs

Character vector. Name of the alternatives to consider in the evaluation. If omitted, all alternatives in performanceTable are used.

criteriaIDs

Character vector. Name of the criteria to consider in the evaluation. If omitted, all criteria in performanceTable are used.

Value

The function returns a vector containing the VIKOR score for each alternative.

References

Opricovic, S. (1998). Multicriteria optimization of civil engineering systems. Faculty of civil engineering, Belgrade, 2(1), 5-21.

Examples

alts <- c("Corsa","Clio","Fiesta")
crit <- c("price","economy", "aesthetics","bootCapacity")
performanceTable <- matrix(c(5490, 51.4, 8.5, 285,
                             6500, 70.6, 7.0, 288,
                             6489, 54.3, 7.5, 290), 
                             nrow=3, ncol=4, byrow=TRUE, 
                             dimnames=list(alts, crit))
criteriaWeights <- setNames(c(0.35,0.25,0.25,0.15), crit)
criteriaMinMax  <- setNames(c("min", "max", "max", "max"), crit)
positiveIdealSolutions <- setNames(c(4500, 80, 9, 300), crit)
negativeIdealSolutions <- setNames(c(7000, 52, 7, 150), crit)

# Overall
VIKOR(performanceTable, criteriaWeights, criteriaMinMax)
# Assuming different ideal and worst solutions
VIKOR(performanceTable, criteriaWeights, criteriaMinMax,
      v=0.5, positiveIdealSolutions, negativeIdealSolutions)
# Using a subset of alternatives and criteria
VIKOR(performanceTable, criteriaWeights, criteriaMinMax,
      v=0.5, positiveIdealSolutions, negativeIdealSolutions,
      alternativesIDs = c("Clio","Fiesta"),
      criteriaIDs = c("price","economy","aesthetics"))