Statistics Seminars: Bayes Linear Variance Adjustment for Dynamic Linear Models with Application to Large Industrial Systems
16 November 2009 14:15 in CM221
Modelling of complex corroding industrial systems is critical to effective inspection and maintenance for assurance of system integrity. We model wall thickness and corrosion rate for multiple dependent corroding components, given observations of minimum wall thickness per component. At each inspection, we do not require that the whole system is observed. We adopt a Bayes Linear approach simplifying parameter estimation and avoiding often-unrealistic distributional assumptions. We also estimate key model variances, making exchangeability assumptions to facilitate analysis for sparse inspection time-series. The model is applied to inspection data from pipework networks on a full-scale offshore platform.
Contact firstname.lastname@example.org for more information