Main objects

DOE

DOE parameters

Parameter	Type	Dimension	Description
Xopt	data.frame	n_estimations × nX	generated DOE points, n_estimations is the sample size defined by the user
Xinfos	object	7 fields	input variable information (check Xinfos parameters for details)
Yinfos	object	7 fields	output variable information (check Yinfos parameters for details)
XY	data.frame	nobs × (nX + nY)	input and output data
X	data.frame	nobs × nX	input data
Y	data.frame	nobs × nY	output data
nobs	numeric	NA	sample size
nX	numeric	NA	number of input variables
nY	numeric	NA	number of output variables
xnames	list	nX elements	input variable names
ynames	list	nY elements	output variable names
xnamesvisu	list	nX elements	input variable names processed for plots
ynamesvisu	list	nY elements	output variable names processed for plots
xnamesmenu	list	nX elements	input variable names processed for menus
ynamesmenu	list	nY elements	output variable names processed for menus
adapt.visu	logical	NA	informs whether the plots should be adapted to the window
idref	list	0 to nobs elements	indices of rows to keep, specified by user selection
idon	list	0 to (nX + nY) elements	indices of columns to keep, specified by user selection
compositeInfos	list of objects	number of composite functions	composite functions information (check compositeInfos object description for details)
Fnames	list	nY elements	functional output variable names
Fnamesvisu	list	nY elements	functional output variable names processed for plots
idF	list	number of functional outputs	vector containing indices of columns for each experimental functional output in DOE$Y
nZ	list	number of functional outputs	vector containing the length of each simulation functional output
discF	list	number of functional outputs	for each functional output, data.frame containing discretization values
Z	data.frame	1 × nY	experimental data values
idZ	list	number of functional outputs	vector containing indices of columns for each simulation functional output in DOE$Y
nZ	list	number of functional outputs	vector containing the length of each experimental functional output
discZ	list	number of functional outputs	for each functional output, data.frame containing discretization values
idZY	list	number of functional outputs	vector containing indices of columns for each functional output, used to map simulation and experimental data when the discretization is different
OF	data.frame	nobs × number of functional outputs	objective function values for each functional output
OFtot	data.frame	nobs × 1	total sum of objective function values

Xinfos parameters

Parameter	Type	Dimension	Description
name	character	NA	variable name
namevisu	character	NA	variable name processed for plots
namemenu	character	NA	variable name processed for menus
type	character	NA	variable type (numeric or categorical)
bounds	list	2 elements	lower bound and upper bound if type is numeric
nlevels	numeric	NA	number of levels if type is categorical
levels	list	nlevels elements	list of levels if type is categorical

Yinfos parameters

Parameter	Type	Dimension	Description
all.ids	list	nY elements	output group (Interest, Control, Status, Constant or Functional)
int.ids	list	0 to nY elements	indices of outputs of interest
control.ids	list	0 to nY elements	indices of control outputs
const.ids	list	0 to nY elements	indices of constant outputs
status.ids	list	0 to nY elements	indices of status outputs
visu.ids	list	0 to nY elements	indices of outputs for visualizations
func.ids	list	0 to nY elements	indices of functional outputs
nY	numeric	NA	number of output variables
type	list	nY elements	output type (numeric or categorical)

compositeInfos object description

Parameter	Type	Dimension	Description
name	character	NA	name of the composite output (entered by the user)
id	numeric	NA	id of the composite output
type	character	NA	variable type (numeric or categorical)
levels	list	1 to n elements	list of levels if type is categorical
usedY	list	1 to n elements	name(s) of the used input(s)
formula	character	NA	formula entered by the user
modelMode	character	NA	informs whether a model should be trained or combine the models from the used parameters (“Train” or “Combine”)
dfNewCol	dataframe	nobs × 1	output result using the current DOE

surrogate

surrogate parameters

Parameter	Type	Dimension	Description
listmodels	object	11 fields	model information (check listmodels parameters for details)
simulations	object	5 fields	simulation information (check simulations parameters for details)

listmodels parameters

Parameter	Type	Dimension	Description
names_surrogatemodel	list	12 elements	names of the available surrogate models
models	list	12 elements	computed models
tableQ2loo	data.frame	12 × (nY + 3)	leave-one-out validation results
tableQ2test	data.frame	12 × (nY + 3)	test validation results
bestQ2loo	list	nY × 2	list of best model for each output with respect to the Q2loo criterion
bestQ2test	list	nY × 2	list of selected model for each output with respect to the Q2test criterion
selected	list	nY × 2	list of selected model for each output, along with the Q2 criterion
trainedModels	list	0 to 12 elements	IDs of trained models
finalpredfun	function	NA	computes prediction with the selected surrogate model
categorical	list	0 to nX elements	indices of categorical input variables
levels.models	list	0 to nX elements	levels of categorical input variables
withsdmodels	list	12 elements	list of boolean values, TRUE if a standard deviation model is available

simulations parameters

Parameter	Type	Dimension	Description
Xadd	data.frame	n_points × nX	extra points to refine the model, n_points is specified by the user
mode.manual	logical	NA	send the additional simulations to the “importDOE” panel where they can be manually launched
mode.automatic	logical	NA	automatically launch the additional simulations with the current simulator settings
tagDOE	character	NA	tag DOE information
nRefine	numeric	NA	refine threshold

UQParams

UQParam parameters

Parameter	Type	Dimension	Description
typeDistr	character	NA	type of distribution
P1Distr	numeric	NA	distribution parameter, depends on typeDistr
P2Distr	numeric	NA	distribution parameter, depends on typeDistr
P3Distr	numeric	NA	distribution parameter, depends on typeDistr
P4Distr	numeric	NA	distribution parameter, depends on typeDistr
levels	list	number of levels	levels for categorical variables
weights	list	number of levels	weights of the levels

calibration

The data structures in this section are specific to the calibration module, they are not used throughout the application. The needed information for the calibration is stored in the object DOE.

calibDOE parameters

Parameter	Type	Dimension	Description
Z	data.frame	1 × nY	experimental data values
sigZ	data.frame	1 × nY	experimental data standard deviations
nZ	list	number of functional outputs	vector containing the length of each experimental functional output
idZ	list	number of functional outputs	vector containing indices of columns for each functional output
idZY	list	number of functional outputs	vector containing indices of columns for each functional output, used to map simulation and experimental data when the discretization is different
discZ	list	number of functional outputs	for each functional output, data.frame containing discretization values
OF	data.frame	nobs × number of functional outputs	objective function values for each functional output
OFtot	data.frame	nobs × 1	total sum of objective function values

objFunc parameters

Parameter	Type	Dimension	Description
norm	character	NA	normalization type (“L1” or “L2”)
weightsTemp	vector	number of functional outputs	weight associated to each functional output (in “Save OF Weights” modal)
weights	vector	number of functional outputs	weight associated to each functional output
OF	data.frame	nobs × number of functional outputs	objective function values for each functional output
OFtot	data.frame	nobs × 1	total sum of objective function values

importDiscF parameters

Parameter	Type	Dimension	Description
discFtemp	data.frame	discretization count × 1	discretization values for some functional outputs in “Import Output Discretization” modal (depends of selected output: “All” or one of the functional outputs)
discF	list	number of functional outputs	for each functional output, data.frame containing discretization values
saveMessage	character	NA	displayed message which indicates discretization saving status

xpData parameters

Parameter	Type	Dimension	Description
Z	data.frame	1 × nY	experimental data values
sigZ	data.frame	1 × nY	experimental data standard deviations
nZ	list	number of functional outputs	vector containing the length of each experimental functional output
idZ	list	number of functional outputs	vector containing indices of columns for each functional output
idZY	list	number of functional outputs	vector containing indices of columns for each functional output, used to map simulation and experimental data when the discretization is different
discZ	list	number of functional outputs	data.frame containing discretization values for each functional output