RMCriteria (version 0.2.0)

SensitivityAnalysis: SensitivityAnalysis

Description

Sensitivity Analysis is a method developed by Wolters & Mareschal (1995) to evaluate how RPrometheeII and RPrometheeIV results are sensitive to changes in weights of criterias. That is, how the solution to the decision problem can be affected by the distribution of criterias weights.

Usage

SensitivityAnalysis(RPrometheeArguments, method = "RPrometheeII")

Arguments

RPrometheeArguments

An object with all RPromethee arguments. For PROMETHEE IV, it's important that parms argument isn't compound of NA values. See RPrometheeConstructor for more information.

method

A character object used to choose how the SensitivityAnalysis is going to be calculated. The method can be "RPrometheeII" or "RPrometheeIV". The standard is "RPrometheeII"

Value

• Solution The solution resulting from linear programming problem.

• alternatives The alternatives names.

• criterias The criterias names.

• datMat The data used corresponding to criterias and alternatives.

References

• M. Behzadian et al. PROMETHEE: A comprehensive literature review on methodologies and applications European Journal of Operational Research v. 200, p.198-215, 2010. https://www.sciencedirect.com/science/article/abs/pii/S0377221709000071

• J. P. Brans, Ph. Vincke A Preference Ranking Organisation Method: (The PROMETHEE Method for Multiple Criteria Decision-Making) Management science, v. 31, n. 6, p. 647-656, 1985. https://pdfs.semanticscholar.org/edd6/f5ae9c1bfb2fdd5c9a5d66e56bdb22770460.pdf

• J. P. Brans, B. Mareschal PROMETHEE methods. In: Figueria J, Greco S, Ehrgott M (eds) Multiple criteria decision analysis: state of the art surveys. Springer Science, Business Media Inc., Boston pp 163???195. http://www.springer.com/la/book/9780387230818

• W.T.M. Wolters, B. Mareschal Novel types of sensitivity analysis for additive MCDM ethods. European Journal of Operational Research, v. 81, p. 281-290, 1995. https://www.sciencedirect.com/science/article/abs/pii/0377221793E0343V

See Also

Other RPromethee methods: PrometheeIIIPlot, PrometheeIIPlot, PrometheeIPlot, PrometheeIVPlot, RPrometheeConstructor, RPrometheeIII, RPrometheeII, RPrometheeIVKernel, RPrometheeIV, RPrometheeI, RPrometheeV, UpdateRPrometheeAlternatives, UpdateRPrometheeArguments, WalkingWeightsPlot, plot,RPrometheeI-method

Examples

# NOT RUN {
## Create objects for each argument
data <- matrix(c(5.2, -3.5,
                 4.3, -1.2,
                 6.7, -2.0,
                 5.4, -5.0,
                 4.8,  0.0,
                 2.8, -3.4), byrow = TRUE, ncol = 2)

parms<-matrix(c(1.0, -2.3), byrow = TRUE, ncol = 1, nrow = 2)
vecWeights <- c(0.3, 0.7)
vecMaximiz <- c(FALSE, TRUE)
prefFunction <- c(0, 0)
constraintDir <- rep("<=", ncol(data))
bounds <- c(7,-1)
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, bounds = bounds,
constraintDir = constraintDir)

## Run RPrometheeV using standard method ("RPrometheeII")
(result <- SensitivityAnalysis(PromObj))

## Run RPrometheeV using RPrometheeIV
(result <- SensitivityAnalysis(PromObj, "RPrometheeIV"))

## Updating alternatives name using UpdateRPrometheeAlternatives
newAlternatives <- c("A", "B", "C", "D", "E", "F")
result <- UpdateRPrometheeAlternatives(result, newAlternatives)

## Updating any argument using UpdateRPrometheeArguments
newParms <- matrix(c(1.6, 4.2), byrow = TRUE, ncol = 1)
PromObj <- UpdateRPrometheeArguments(PromObj, "parms", newParms)
(result <- SensitivityAnalysis(PromObj))
# }