Penalized likelihood kriging model for analysis of computer experiments

Runze Li, Agus Sudjianto

Research output: Contribution to conferencePaper

7 Citations (Scopus)

Abstract

Kriging is a popular metamodeling technique for analysis of computer experiment. However, the likelihood function near the optimum is flat in some situations, and this may lead to very large random variation in the maximum likelihood estimate. To overcome this difficulty, a penalized likelihood approach is proposed for the kriging model. The proposed method is particularly important in the context of a computationally intensive simulation model where the number of simulation runs must be kept small. We demonstrate the proposed approach for the reduction of piston slap, an unwanted engine noise due to piston secondary motion. Issues related to practical implementation of the proposed approach are discussed.

Original languageEnglish (US)
Pages527-533
Number of pages7
StatePublished - Dec 1 2003
Event2003 ASME Design Engineering Technical Conferences and Computers and Information in Engineering Conference - Chicago, IL, United States
Duration: Sep 2 2003Sep 6 2003

Other

Other2003 ASME Design Engineering Technical Conferences and Computers and Information in Engineering Conference
CountryUnited States
CityChicago, IL
Period9/2/039/6/03

Fingerprint

Penalized Likelihood
Computer Experiments
Kriging
Pistons
Metamodeling
Likelihood Function
Maximum Likelihood Estimate
Maximum likelihood
Simulation Model
Engine
Experiments
Engines
Motion
Model
Demonstrate
Simulation
Context

All Science Journal Classification (ASJC) codes

  • Modeling and Simulation
  • Mechanical Engineering
  • Computer Science Applications
  • Computer Graphics and Computer-Aided Design

Cite this

Li, R., & Sudjianto, A. (2003). Penalized likelihood kriging model for analysis of computer experiments. 527-533. Paper presented at 2003 ASME Design Engineering Technical Conferences and Computers and Information in Engineering Conference, Chicago, IL, United States.
Li, Runze ; Sudjianto, Agus. / Penalized likelihood kriging model for analysis of computer experiments. Paper presented at 2003 ASME Design Engineering Technical Conferences and Computers and Information in Engineering Conference, Chicago, IL, United States.7 p.
@conference{c0ce1289de624fb898c852bfd6078a08,
title = "Penalized likelihood kriging model for analysis of computer experiments",
abstract = "Kriging is a popular metamodeling technique for analysis of computer experiment. However, the likelihood function near the optimum is flat in some situations, and this may lead to very large random variation in the maximum likelihood estimate. To overcome this difficulty, a penalized likelihood approach is proposed for the kriging model. The proposed method is particularly important in the context of a computationally intensive simulation model where the number of simulation runs must be kept small. We demonstrate the proposed approach for the reduction of piston slap, an unwanted engine noise due to piston secondary motion. Issues related to practical implementation of the proposed approach are discussed.",
author = "Runze Li and Agus Sudjianto",
year = "2003",
month = "12",
day = "1",
language = "English (US)",
pages = "527--533",
note = "2003 ASME Design Engineering Technical Conferences and Computers and Information in Engineering Conference ; Conference date: 02-09-2003 Through 06-09-2003",

}

Li, R & Sudjianto, A 2003, 'Penalized likelihood kriging model for analysis of computer experiments' Paper presented at 2003 ASME Design Engineering Technical Conferences and Computers and Information in Engineering Conference, Chicago, IL, United States, 9/2/03 - 9/6/03, pp. 527-533.

Penalized likelihood kriging model for analysis of computer experiments. / Li, Runze; Sudjianto, Agus.

2003. 527-533 Paper presented at 2003 ASME Design Engineering Technical Conferences and Computers and Information in Engineering Conference, Chicago, IL, United States.

Research output: Contribution to conferencePaper

TY - CONF

T1 - Penalized likelihood kriging model for analysis of computer experiments

AU - Li, Runze

AU - Sudjianto, Agus

PY - 2003/12/1

Y1 - 2003/12/1

N2 - Kriging is a popular metamodeling technique for analysis of computer experiment. However, the likelihood function near the optimum is flat in some situations, and this may lead to very large random variation in the maximum likelihood estimate. To overcome this difficulty, a penalized likelihood approach is proposed for the kriging model. The proposed method is particularly important in the context of a computationally intensive simulation model where the number of simulation runs must be kept small. We demonstrate the proposed approach for the reduction of piston slap, an unwanted engine noise due to piston secondary motion. Issues related to practical implementation of the proposed approach are discussed.

AB - Kriging is a popular metamodeling technique for analysis of computer experiment. However, the likelihood function near the optimum is flat in some situations, and this may lead to very large random variation in the maximum likelihood estimate. To overcome this difficulty, a penalized likelihood approach is proposed for the kriging model. The proposed method is particularly important in the context of a computationally intensive simulation model where the number of simulation runs must be kept small. We demonstrate the proposed approach for the reduction of piston slap, an unwanted engine noise due to piston secondary motion. Issues related to practical implementation of the proposed approach are discussed.

UR - http://www.scopus.com/inward/record.url?scp=1842811035&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=1842811035&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:1842811035

SP - 527

EP - 533

ER -

Li R, Sudjianto A. Penalized likelihood kriging model for analysis of computer experiments. 2003. Paper presented at 2003 ASME Design Engineering Technical Conferences and Computers and Information in Engineering Conference, Chicago, IL, United States.