# Kriging: Methods and Applications

Research output: Working paperDiscussion paperOther research output

### Abstract

In this chapter we present Kriging— also known as a Gaussian process (GP) model— which is a mathematical interpolation method. To select the input combinations to be simulated, we use Latin hypercube sampling (LHS); we allow uniform and non-uniform distributions of the simulation inputs. Besides deterministic simulation we discuss random simulation, which requires adjusting the design and analysis. We discuss sensitivity analysis of simulation models, using "functional analysis of variance" (FANOVA)— also known as Sobol sensitivity indexes. Finally, we discuss
optimization of the simulated system, including "robust" optimization.
Original language English Tilburg CentER, Center for Economic Research 17 2017-047 Published - 21 Nov 2017

### Publication series

Name CentER Discussion Paper 2017-047

### Fingerprint

Kriging
Latin Hypercube Sampling
Simulation
Robust Optimization
Interpolation Method
Functional Analysis
Gaussian Model
Analysis of variance
Model Analysis
Gaussian Process
Process Model
Sensitivity Analysis
Simulation Model

### Keywords

• Gaussian process
• Latin hypercube
• deterministic simulation
• random simulation
• sensitivity analysis
• optimization

### Cite this

Kleijnen, J. P. C. (2017). Kriging: Methods and Applications. (CentER Discussion Paper; Vol. 2017-047). Tilburg: CentER, Center for Economic Research.
Kleijnen, J.P.C. / Kriging : Methods and Applications. Tilburg : CentER, Center for Economic Research, 2017. (CentER Discussion Paper).
@techreport{0e31d8b9596f4bbca24800b22c757ba9,
title = "Kriging: Methods and Applications",
abstract = "In this chapter we present Kriging— also known as a Gaussian process (GP) model— which is a mathematical interpolation method. To select the input combinations to be simulated, we use Latin hypercube sampling (LHS); we allow uniform and non-uniform distributions of the simulation inputs. Besides deterministic simulation we discuss random simulation, which requires adjusting the design and analysis. We discuss sensitivity analysis of simulation models, using {"}functional analysis of variance{"} (FANOVA)— also known as Sobol sensitivity indexes. Finally, we discussoptimization of the simulated system, including {"}robust{"} optimization.",
keywords = "Gaussian process, Latin hypercube, deterministic simulation, random simulation, sensitivity analysis, optimization",
author = "J.P.C. Kleijnen",
year = "2017",
month = "11",
day = "21",
language = "English",
volume = "2017-047",
series = "CentER Discussion Paper",
publisher = "CentER, Center for Economic Research",
type = "WorkingPaper",
institution = "CentER, Center for Economic Research",

}

Kleijnen, JPC 2017 'Kriging: Methods and Applications' CentER Discussion Paper, vol. 2017-047, CentER, Center for Economic Research, Tilburg.
Tilburg : CentER, Center for Economic Research, 2017. (CentER Discussion Paper; Vol. 2017-047).

Research output: Working paperDiscussion paperOther research output

TY - UNPB

T1 - Kriging

T2 - Methods and Applications

AU - Kleijnen, J.P.C.

PY - 2017/11/21

Y1 - 2017/11/21

N2 - In this chapter we present Kriging— also known as a Gaussian process (GP) model— which is a mathematical interpolation method. To select the input combinations to be simulated, we use Latin hypercube sampling (LHS); we allow uniform and non-uniform distributions of the simulation inputs. Besides deterministic simulation we discuss random simulation, which requires adjusting the design and analysis. We discuss sensitivity analysis of simulation models, using "functional analysis of variance" (FANOVA)— also known as Sobol sensitivity indexes. Finally, we discussoptimization of the simulated system, including "robust" optimization.

AB - In this chapter we present Kriging— also known as a Gaussian process (GP) model— which is a mathematical interpolation method. To select the input combinations to be simulated, we use Latin hypercube sampling (LHS); we allow uniform and non-uniform distributions of the simulation inputs. Besides deterministic simulation we discuss random simulation, which requires adjusting the design and analysis. We discuss sensitivity analysis of simulation models, using "functional analysis of variance" (FANOVA)— also known as Sobol sensitivity indexes. Finally, we discussoptimization of the simulated system, including "robust" optimization.

KW - Gaussian process

KW - Latin hypercube

KW - deterministic simulation

KW - random simulation

KW - sensitivity analysis

KW - optimization

M3 - Discussion paper

VL - 2017-047

T3 - CentER Discussion Paper

BT - Kriging

PB - CentER, Center for Economic Research

CY - Tilburg

ER -

Kleijnen JPC. Kriging: Methods and Applications. Tilburg: CentER, Center for Economic Research. 2017 Nov 21. (CentER Discussion Paper).