screeningIntervals

fun screeningIntervals(groups: List<EstimatedResponseIfc>, probCS: Double = DEFAULT_CONFIDENCE_LEVEL, indifferenceZone: Double = 0.0, optimizationType: OptimizationType = OptimizationType.MINIMIZE): Map<Int, Map<Int, Interval>>(source)

Returns the screening intervals for each alternative assuming the specified direction of optimization. Returns a map of maps. The keys for the maps are the indices of the alternatives (in combination). The interval represents the screening interval. The intervals are one-sided negative infinity, upper limit for minimization and lower limit, positive infinity for maximization. The intervals can be checked to determine if the alternative should be retained after screening.

Based on: Nelson et al. (2001) “Simple Procedures for Selecting the Best System when the Number of Alternatives is Large,” Operations Research, vol. 49, pp.950-963.

Parameters

groups

the groups to be screened. The group's size must be 2 or more. The sample size for each group must be two or more.

probCS

is the probability of correct selection for the screening.

indifferenceZone

the indifference zone parameter. If the systems are within this value, then we are indifferent to the difference. The default is 0.0.

optimizationType

the type of optimization {MINIMIZE or MAXIMIZE}. The default is MINIMIZE.