MCDA (version 0.1.0)

SURE: Simulated Uncertainty Range Evaluations (SURE)

Description

SURE is a multi-criteria decision analysis method which was developed by Richard Hodgett and Sajid Siraj. More details on the method are available in https://doi.org/10.1016/j.eswa.2018.08.048

Usage

SURE(
  performanceTableMin,
  performanceTable,
  performanceTableMax,
  criteriaWeights,
  criteriaMinMax,
  alternativesIDs = NULL,
  criteriaIDs = NULL,
  NoOfSimulations = 1e+05
)

Value

The function returns an element of type SURE which contains the SURE simulated scores for each alternative.

Arguments

performanceTableMin

Matrix or data frame containing the minimum performance table. Each column corresponds to an alternative, and each row to a criterion. Columns (resp. rows) must be named according to the IDs of the alternatives (resp. criteria).

performanceTable

Matrix or data frame containing the most likely performance table. Each column corresponds to an alternative, and each row to a criterion. Columns (resp. rows) must be named according to the IDs of the alternatives (resp. criteria).

performanceTableMax

Matrix or data frame containing the maximum performance table. Each column corresponds to an alternative, and each row to a criterion. Columns (resp. rows) must be named according to the IDs of the alternatives (resp. criteria).

criteriaWeights

Vector containing the weights of the criteria. The elements are named according to the IDs of the criteria.

criteriaMinMax

Vector containing the preference direction on each of the criteria. "min" (resp. "max") indicates that the criterion has to be minimized (maximized). The elements are named according to the IDs of the criteria.

alternativesIDs

Vector containing IDs of alternatives, according to which the data should be filtered.

criteriaIDs

Vector containing IDs of criteria, according to which the data should be filtered.

NoOfSimulations

Integer stating the number of Simulations to use.

References

Richard E. Hodgett, Sajid Siraj (2019). SURE: A method for decision-making under uncertainty. Expert Systems with Applications, Volume 115, 684-694.

Examples

performanceTableMin <- t(matrix(c(78,87,79,19,8,68,74,8,90,89,74.5,9,20,81,30),
                  nrow=3,ncol=5, byrow=TRUE)) 
performanceTable <- t(matrix(c(80,87,86,19,8,70,74,10,90,89,75,9,33,82,30),
                              nrow=3,ncol=5, byrow=TRUE))
performanceTableMax <- t(matrix(c(81,87,95,19,8,72,74,15,90,89,75.5,9,36,84,30),
                                 nrow=3,ncol=5, byrow=TRUE))  

row.names(performanceTable) <- c("Yield","Toxicity","Cost","Separation","Odour")
colnames(performanceTable) <- c("Route One","Route Two","Route Three")
row.names(performanceTableMin) <- row.names(performanceTable)
colnames(performanceTableMin) <- colnames(performanceTable)
row.names(performanceTableMax) <- row.names(performanceTable)
colnames(performanceTableMax) <- colnames(performanceTable)

criteriaWeights <- c(0.339,0.077,0.434,0.127,0.023) 
names(criteriaWeights) <- row.names(performanceTable)

criteriaMinMax <- c("max", "max", "max", "max", "max")
names(criteriaMinMax) <- row.names(performanceTable)

test1 <- SURE(performanceTableMin, 
                 performanceTable, 
                 performanceTableMax, 
                 criteriaWeights, 
                 criteriaMinMax, NoOfSimulations = 101)

summary(test1)
plotSURE(test1)
plotSURE(test1, greyScale = TRUE, separate = TRUE)

test2 <- SURE(performanceTableMin, 
              performanceTable,
              performanceTableMax,
              criteriaWeights,
              criteriaMinMax,
              alternativesIDs = c("Route Two","Route Three"),
              criteriaIDs = c("Yield","Toxicity","Separation"),
              NoOfSimulations = 101)

summary(test2)
plotSURE(test2)
plotSURE(test2, greyScale = TRUE, separate = TRUE)