Learn R Programming

LatticeDesign (version 4.0-1)

LLHD: Optimal generator vector for Lattice-based Latin hypercube designs

Description

Generates the optimal generator vector for lattice-based Latin hypercube designs.

Usage

LLHD(n,d,criterion="WS",T=1000,nstart=max(c(floor(T/(length(CoprimeNumbers(n))*d*5)),1)));

Value

The value returned from the function gives a generator vector for lattice-based Latin hypercube designs.

Arguments

n

Number of points, must be an integer greater than one.

d

Number of dimensions, must be an integer greater than one.

criterion

Optional, the criterion used in optimization, which can be "WS", "WP", "WD", "bWS", "bWF", or "bWD".

T

Optional, number of iterations.

nstart

Optional, number of random starts.

Details

This function generates the optimal generator vector for lattice-based Latin hypercube designs, following the algorithms provided in the paper "optimized and regularly repeated lattice-based Latin hypercube designs for large-scale computer experiment".

References

Xu He, Junpeng Gong, and Zhaohui Li (2025) "optimized and regularly repeated lattice-based Latin hypercube designs for large-scale computer experiment", <arXiv:2506.04582v1>

Examples

Run this code
set.seed(85)
v <- LLHD(n=20,d=3,criterion="WS")
design <- LLHDpoints(n=20,v=v,delta=sample(x=1:20,size=3,replace=TRUE))
design

Run the code above in your browser using DataLab