Uncertainty quantification via elicitation of expert judgements
Uncertainty quantification via elicitation of expert judgements
The purpose of this paper is to depict one method of quantifying uncertainty about different parameters, which is based on eliciting judgements either from a single expert or from a group of experts. The quantities obtained as a result of the elicitation are therefore used in order to fit probability density functions (PDFs) by using an in-house MATLAB model which uses appropriate fitting techniques similar to the ones suggested in the existing literature. Consequently, an initial framework has been implemented which would first of all allow the comparison of elicited data with the experimental results. The underlying theory behind the elicitation process is being presented and subsequently an aero-engine Fan Blade Off (FBO) case study is presented. The framework is used to illustrate the way in which expert judgements are implemented as inputs into the MATLAB model which is used to predict different parameters of interest associated to FBO events such as probabilities of having a particular speed during an event as well as what are the characteristics of the most likely events to occur. Those are taken into consideration in order to allow the designer to perform relevant and more detailed analysis on the fan subsystem during the preliminary design process.
Profir, Bogdan
6db80893-c830-4dbc-87bc-4c0a15077d06
Eres, Hakki
b22e2d66-55c4-46d2-8ec3-46317033de43
Scanlan, James
7ad738f2-d732-423f-a322-31fa4695529d
Moss, Michael
4cc758f8-0e77-4a92-a953-79f76b4e69c0
Bates, Ron
f3439cad-2150-43de-8513-d5fc90317be7
Profir, Bogdan
6db80893-c830-4dbc-87bc-4c0a15077d06
Eres, Hakki
b22e2d66-55c4-46d2-8ec3-46317033de43
Scanlan, James
7ad738f2-d732-423f-a322-31fa4695529d
Moss, Michael
4cc758f8-0e77-4a92-a953-79f76b4e69c0
Bates, Ron
f3439cad-2150-43de-8513-d5fc90317be7
Profir, Bogdan, Eres, Hakki, Scanlan, James, Moss, Michael and Bates, Ron
(2016)
Uncertainty quantification via elicitation of expert judgements.
17th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference, , Washington D.C., United States.
13 - 17 Jun 2016.
(In Press)
Record type:
Conference or Workshop Item
(Paper)
Abstract
The purpose of this paper is to depict one method of quantifying uncertainty about different parameters, which is based on eliciting judgements either from a single expert or from a group of experts. The quantities obtained as a result of the elicitation are therefore used in order to fit probability density functions (PDFs) by using an in-house MATLAB model which uses appropriate fitting techniques similar to the ones suggested in the existing literature. Consequently, an initial framework has been implemented which would first of all allow the comparison of elicited data with the experimental results. The underlying theory behind the elicitation process is being presented and subsequently an aero-engine Fan Blade Off (FBO) case study is presented. The framework is used to illustrate the way in which expert judgements are implemented as inputs into the MATLAB model which is used to predict different parameters of interest associated to FBO events such as probabilities of having a particular speed during an event as well as what are the characteristics of the most likely events to occur. Those are taken into consideration in order to allow the designer to perform relevant and more detailed analysis on the fan subsystem during the preliminary design process.
Text
Bogdan Profir - AIAA 2016 Accepted Version.pdf
- Accepted Manuscript
Restricted to Repository staff only
Request a copy
More information
Accepted/In Press date: 17 February 2016
Venue - Dates:
17th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference, , Washington D.C., United States, 2016-06-13 - 2016-06-17
Organisations:
Computational Engineering & Design Group
Identifiers
Local EPrints ID: 393987
URI: http://eprints.soton.ac.uk/id/eprint/393987
PURE UUID: a0901de2-25a4-4640-abd1-1c94421c8652
Catalogue record
Date deposited: 01 Jun 2016 11:23
Last modified: 15 Mar 2024 03:14
Export record
Contributors
Author:
Bogdan Profir
Author:
Michael Moss
Author:
Ron Bates
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
