Input model uncertainty assessment: A study within the automotive industry
Input model uncertainty assessment: A study within the automotive industry
Input model uncertainty refers to the uncertainty surrounding the choice of distributions and their parameters, due to the use of finite samples from the population. Input model uncertainty is often not included in the standard output analysis, something that could result in confidence intervals that are too optimistic. This paper discusses how the input model uncertainty in a model used by Ford Motor Company is quantified using mean-variance metamodel approximation. The variance caused by input model uncertainty is deduced and expressed in units of simulation sampling error. The assessment estimates the distributions’ contributions to input uncertainty and the sample size sensitivities. The method also entails the construction of a metamodel that relates the means and variances of the distributions included in the assessment, to the means of the simulation output. This metamodel, could be used as a quick stand-in to the model comprising of the distributions included in the assessment.
98-104
Operational Research Society
Ioannidis, P.
eb2357ae-df18-46aa-98cd-7eb833f43707
Onggo, B.S.
8e9a2ea5-140a-44c0-9c17-e9cf93662f80
Higgins, M.
e62fcdc1-e238-4790-8d43-985fc8edc2b3
Ladbrook, J.
c88fc9fe-87e1-459a-a1d6-c8bb8cbcd93b
2017
Ioannidis, P.
eb2357ae-df18-46aa-98cd-7eb833f43707
Onggo, B.S.
8e9a2ea5-140a-44c0-9c17-e9cf93662f80
Higgins, M.
e62fcdc1-e238-4790-8d43-985fc8edc2b3
Ladbrook, J.
c88fc9fe-87e1-459a-a1d6-c8bb8cbcd93b
Ioannidis, P., Onggo, B.S., Higgins, M. and Ladbrook, J.
(2017)
Input model uncertainty assessment: A study within the automotive industry.
Robertson, D., Fakhimi, M., Anagnostou, A. and Meskarian, R.
(eds.)
In Proceedings of the Operational Research Society Simulation Workshop 2018.
Operational Research Society.
.
Record type:
Conference or Workshop Item
(Paper)
Abstract
Input model uncertainty refers to the uncertainty surrounding the choice of distributions and their parameters, due to the use of finite samples from the population. Input model uncertainty is often not included in the standard output analysis, something that could result in confidence intervals that are too optimistic. This paper discusses how the input model uncertainty in a model used by Ford Motor Company is quantified using mean-variance metamodel approximation. The variance caused by input model uncertainty is deduced and expressed in units of simulation sampling error. The assessment estimates the distributions’ contributions to input uncertainty and the sample size sensitivities. The method also entails the construction of a metamodel that relates the means and variances of the distributions included in the assessment, to the means of the simulation output. This metamodel, could be used as a quick stand-in to the model comprising of the distributions included in the assessment.
This record has no associated files available for download.
More information
Published date: 2017
Additional Information:
Export Date: 9 October 2018
Identifiers
Local EPrints ID: 425185
URI: http://eprints.soton.ac.uk/id/eprint/425185
PURE UUID: 85fab462-0816-49ff-bd2d-967aa209210b
Catalogue record
Date deposited: 11 Oct 2018 16:30
Last modified: 08 Jan 2022 03:38
Export record
Contributors
Author:
P. Ioannidis
Author:
M. Higgins
Author:
J. Ladbrook
Editor:
D. Robertson
Editor:
M. Fakhimi
Editor:
A. Anagnostou
Editor:
R. Meskarian
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics
