Home > Research > Publications & Outputs > Bayes linear variance structure learning for in...

Research

Links

Text available via DOI:

Keywords

View graph of relations

Bayes linear variance structure learning for inspection of large scale physical systems

Research output: Contribution to Journal/Magazine › Journal article › peer-review

Published

Standard

Bayes linear variance structure learning for inspection of large scale physical systems. / Randell, D.; Goldstein, M.; Jonathan, P.
In: Proceedings of the Institution of Mechanical Engineers, Part O: Journal of Risk and Reliability, Vol. 228, No. 1, 2014, p. 3-18.

Research output: Contribution to Journal/Magazine › Journal article › peer-review

Harvard

Randell, D, Goldstein, M & Jonathan, P 2014, 'Bayes linear variance structure learning for inspection of large scale physical systems', Proceedings of the Institution of Mechanical Engineers, Part O: Journal of Risk and Reliability, vol. 228, no. 1, pp. 3-18. https://doi.org/10.1177/1748006X13492955

APA

Randell, D., Goldstein, M., & Jonathan, P. (2014). Bayes linear variance structure learning for inspection of large scale physical systems. Proceedings of the Institution of Mechanical Engineers, Part O: Journal of Risk and Reliability, 228(1), 3-18. https://doi.org/10.1177/1748006X13492955

Vancouver

Randell D, Goldstein M, Jonathan P. Bayes linear variance structure learning for inspection of large scale physical systems. Proceedings of the Institution of Mechanical Engineers, Part O: Journal of Risk and Reliability. 2014;228(1):3-18. doi: 10.1177/1748006X13492955

Author

Randell, D. ; Goldstein, M. ; Jonathan, P. / Bayes linear variance structure learning for inspection of large scale physical systems. In: Proceedings of the Institution of Mechanical Engineers, Part O: Journal of Risk and Reliability. 2014 ; Vol. 228, No. 1. pp. 3-18.

Bibtex

@article{2124551a422e43e3a397db4b83550338,
title = "Bayes linear variance structure learning for inspection of large scale physical systems",
abstract = "Modelling of inspection data for large scale physical systems is critical to assessment of their integrity. We present a general method for inference about system state and associated model variance structure from spatially distributed time series that are typically short, irregular, incomplete and not directly observable. Bayes linear analysis simplifies parameter estimation and avoids often-unrealistic distributional assumptions. Second-order exchangeability judgements facilitate variance learning for sparse inspection time-series. The model is applied to inspection data for minimum wall thickness from corroding pipe-work networks on a full-scale offshore platform, and shown to give materially different forecasts of remnant life compared with an equivalent model neglecting variance learning. {\textcopyright} 2013 IMechE.",
keywords = "Bayes linear, corrosion, dynamic linear model, exchangeability, Mahalanobis distance, variance learning, Dynamic linear model, Mahalanobis distances, Corrosion, Drilling platforms, Offshore structures, Inspection",
author = "D. Randell and M. Goldstein and P. Jonathan",
year = "2014",
doi = "10.1177/1748006X13492955",
language = "English",
volume = "228",
pages = "3--18",
journal = "Proceedings of the Institution of Mechanical Engineers, Part O: Journal of Risk and Reliability",
issn = "1748-006X",
publisher = "SAGE Publications Ltd",
number = "1",

}

RIS

TY - JOUR

T1 - Bayes linear variance structure learning for inspection of large scale physical systems

AU - Randell, D.

AU - Goldstein, M.

AU - Jonathan, P.

PY - 2014

Y1 - 2014

N2 - Modelling of inspection data for large scale physical systems is critical to assessment of their integrity. We present a general method for inference about system state and associated model variance structure from spatially distributed time series that are typically short, irregular, incomplete and not directly observable. Bayes linear analysis simplifies parameter estimation and avoids often-unrealistic distributional assumptions. Second-order exchangeability judgements facilitate variance learning for sparse inspection time-series. The model is applied to inspection data for minimum wall thickness from corroding pipe-work networks on a full-scale offshore platform, and shown to give materially different forecasts of remnant life compared with an equivalent model neglecting variance learning. © 2013 IMechE.

AB - Modelling of inspection data for large scale physical systems is critical to assessment of their integrity. We present a general method for inference about system state and associated model variance structure from spatially distributed time series that are typically short, irregular, incomplete and not directly observable. Bayes linear analysis simplifies parameter estimation and avoids often-unrealistic distributional assumptions. Second-order exchangeability judgements facilitate variance learning for sparse inspection time-series. The model is applied to inspection data for minimum wall thickness from corroding pipe-work networks on a full-scale offshore platform, and shown to give materially different forecasts of remnant life compared with an equivalent model neglecting variance learning. © 2013 IMechE.

KW - Bayes linear

KW - corrosion

KW - dynamic linear model

KW - exchangeability

KW - Mahalanobis distance

KW - variance learning

KW - Dynamic linear model

KW - Mahalanobis distances

KW - Corrosion

KW - Drilling platforms

KW - Offshore structures

KW - Inspection

U2 - 10.1177/1748006X13492955

DO - 10.1177/1748006X13492955

M3 - Journal article

VL - 228

SP - 3

EP - 18

JO - Proceedings of the Institution of Mechanical Engineers, Part O: Journal of Risk and Reliability

JF - Proceedings of the Institution of Mechanical Engineers, Part O: Journal of Risk and Reliability

SN - 1748-006X

IS - 1

ER -