Robust Optimization in Simulation: Taguchi and Krige Combined

G. Dellino, Jack P.C. Kleijnen, C. Meloni

Research output: Working paperDiscussion paperOther research output

379 Downloads (Pure)

Abstract

Optimization of simulated systems is the goal of many methods, but most methods as- sume known environments. We, however, develop a `robust' methodology that accounts for uncertain environments. Our methodology uses Taguchi's view of the uncertain world, but replaces his statistical techniques by Kriging. We illustrate the resulting methodology through classic Economic Order Quantity (EOQ) inventory models. Our results suggest that robust optimization requires order quantities that dier from the classic EOQ. We also compare our latest results with our previous results that do not use Kriging but Response Surface Methodology (RSM).
Original languageEnglish
Place of PublicationTilburg
PublisherOperations research
Number of pages29
Volume2009-82
Publication statusPublished - 2009

Publication series

NameCentER Discussion Paper
Volume2009-82

Fingerprint

Economics

Keywords

  • Statistics
  • Design of experiments
  • Inventory-Production
  • Simulation
  • Decision analysis

Cite this

Dellino, G., Kleijnen, J. P. C., & Meloni, C. (2009). Robust Optimization in Simulation: Taguchi and Krige Combined. (CentER Discussion Paper; Vol. 2009-82). Tilburg: Operations research.
Dellino, G. ; Kleijnen, Jack P.C. ; Meloni, C. / Robust Optimization in Simulation : Taguchi and Krige Combined. Tilburg : Operations research, 2009. (CentER Discussion Paper).
@techreport{d919b893db2b4d97a3924a55e2ad2494,
title = "Robust Optimization in Simulation: Taguchi and Krige Combined",
abstract = "Optimization of simulated systems is the goal of many methods, but most methods as- sume known environments. We, however, develop a `robust' methodology that accounts for uncertain environments. Our methodology uses Taguchi's view of the uncertain world, but replaces his statistical techniques by Kriging. We illustrate the resulting methodology through classic Economic Order Quantity (EOQ) inventory models. Our results suggest that robust optimization requires order quantities that dier from the classic EOQ. We also compare our latest results with our previous results that do not use Kriging but Response Surface Methodology (RSM).",
keywords = "Statistics, Design of experiments, Inventory-Production, Simulation, Decision analysis",
author = "G. Dellino and Kleijnen, {Jack P.C.} and C. Meloni",
note = "Subsequently published in Informs Journal on Computing (2012) Pagination: 29",
year = "2009",
language = "English",
volume = "2009-82",
series = "CentER Discussion Paper",
publisher = "Operations research",
type = "WorkingPaper",
institution = "Operations research",

}

Dellino, G, Kleijnen, JPC & Meloni, C 2009 'Robust Optimization in Simulation: Taguchi and Krige Combined' CentER Discussion Paper, vol. 2009-82, Operations research, Tilburg.

Robust Optimization in Simulation : Taguchi and Krige Combined. / Dellino, G.; Kleijnen, Jack P.C.; Meloni, C.

Tilburg : Operations research, 2009. (CentER Discussion Paper; Vol. 2009-82).

Research output: Working paperDiscussion paperOther research output

TY - UNPB

T1 - Robust Optimization in Simulation

T2 - Taguchi and Krige Combined

AU - Dellino, G.

AU - Kleijnen, Jack P.C.

AU - Meloni, C.

N1 - Subsequently published in Informs Journal on Computing (2012) Pagination: 29

PY - 2009

Y1 - 2009

N2 - Optimization of simulated systems is the goal of many methods, but most methods as- sume known environments. We, however, develop a `robust' methodology that accounts for uncertain environments. Our methodology uses Taguchi's view of the uncertain world, but replaces his statistical techniques by Kriging. We illustrate the resulting methodology through classic Economic Order Quantity (EOQ) inventory models. Our results suggest that robust optimization requires order quantities that dier from the classic EOQ. We also compare our latest results with our previous results that do not use Kriging but Response Surface Methodology (RSM).

AB - Optimization of simulated systems is the goal of many methods, but most methods as- sume known environments. We, however, develop a `robust' methodology that accounts for uncertain environments. Our methodology uses Taguchi's view of the uncertain world, but replaces his statistical techniques by Kriging. We illustrate the resulting methodology through classic Economic Order Quantity (EOQ) inventory models. Our results suggest that robust optimization requires order quantities that dier from the classic EOQ. We also compare our latest results with our previous results that do not use Kriging but Response Surface Methodology (RSM).

KW - Statistics

KW - Design of experiments

KW - Inventory-Production

KW - Simulation

KW - Decision analysis

M3 - Discussion paper

VL - 2009-82

T3 - CentER Discussion Paper

BT - Robust Optimization in Simulation

PB - Operations research

CY - Tilburg

ER -

Dellino G, Kleijnen JPC, Meloni C. Robust Optimization in Simulation: Taguchi and Krige Combined. Tilburg: Operations research. 2009. (CentER Discussion Paper).