wtLHDs function

Weighted Criteria-Based Latin Hypercube Designs (LHDs) for Any Numbers of Factors (>=2)

For any number of factors, F (>=2), this method will generate LHDs with levels, L ranges from F to sC2, where s=2*F+1. Maxpro criterion measure, Phi_p measure also provided as a measure of space-filling and as an orthogonality measure, maximum absolute correlation (MAC) value also provided.

wtLHDs(levels,factors,weights,iterations,population)

Arguments

• levels: Ranges from F to sC2, where s=2*F+1
• factors: F, any number (>=2)
• weights: Weight should be given to MAC, Phi_p and Maxpro such that sum is 1. Default it is 0.2, 0.4 and 0.4
• iterations: Number of iterations. By default it is 300
• population: Population size. By default it is 100

Returns

This function will provide a series of weighted criteria-based LHDs along with space-filling and orthogonality measures for the generated LHDs.

Author(s)

Ashutosh Dalal, Cini Varghese, Rajender Parsad and Mohd Harun

Examples

library(CompExpDes)
wtLHDs(20,3,c(0.4,0.2,0.4))

References

McKay, M.D., Beckman, R.J. and Conover, W.J. (1979). Comparison of three methods for selecting values of input variables in the analysis of output from a computer code. Technometrics, 21(2), 239-245.