This function generates a MaxPro Latin-hypercube design.
maxproLHD(
n,
p,
design = NULL,
max.sa.iter = 1e+06,
temp = 0,
decay = 0.95,
no.update.iter.max = 400,
num.passes = 10,
max.det.iter = 1e+06,
method = "full",
scaled = TRUE
)final design points.
total number of swaps in the optimization.
final optimized criterion.
criterion history during the optimization process.
design size.
design dimension.
an initial LHD. If design=NULL, a random LHD is generated.
maximum number of swapping involved in the simulated annealing (SA) algorithm.
initial temperature of the simulated annealing algorithm. If temp=0, it will be automatically determined.
the temperature decay rate of simulated annealing.
the maximum number of iterations where there is no update to the global optimum before SA stops.
the maximum number of passes of the whole design matrix if deterministic swapping is used.
maximum number of swapping involved in the deterministic swapping algorithm.
choice of "deterministic", "sa", or "full". If the method="full", the design is first optimized by SA and then deterministic swapping.
whether the design is scaled to unit hypercube. If scaled=FALSE, the design is represented by integer numbers from 1 to design size. Leave it as TRUE when no initial design is provided.
maxproLHD generates a MaxPro Latin-hypercube design (Joseph, V. R., Gul, E., & Ba, S. 2015). The major difference with the MaxPro packages is that we have a deterministic swap algorithm, which can be enabled by setting method="deterministic" or method="full". For optimization details, see the detail section in customLHD.
Joseph, V. R., Gul, E., & Ba, S. (2015). Maximum projection designs for computer experiments. Biometrika, 102(2), 371-380.
n = 20
p = 3
D = maxproLHD(n, p)
Run the code above in your browser using DataLab