Bayesian linear inspection planning for large-scale physical systems

Abstract

Modelling of complex corroding industrial systems is critical to effective inspection and maintenance for assurance of system integrity. Wall thickness and corrosion rate are modelled for multiple dependent corroding components, given observations of minimum wall thickness per component. At each inspection, partial observations of the system are considered. A Bayes Linear approach is adopted simplifying parameter estimation and avoiding often unrealistic distributional assumptions. Key system variances are modelled, making exchangeability assumptions to facilitate analysis for sparse inspection time-series. A utility based criterion is used to assess quality of inspection design and aid decision making. The model is applied to inspection data from pipework networks on a full-scale offshore platform.

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

“…A corroding system may exhibit multiple corrosion circuits with similar corrosion behaviour. This application was discussed previously by Randell et al (2010).…”

“…A corroding system may exhibit multiple corrosion circuits with similar corrosion behaviour. This application was discussed previously by Randell et al 11 When components are inspected, inspectors are most concerned with the most vulnerable components since these are critical to assessing the risk of system failure. Measurements are made using non-destructive (e.g.…”

“…Shaw and Goldstein (1999) discusses Bayes linear experimental design for grouped multivariate exchangeable systems. Randell et al (2010) develops a utility-based criterion to assess inspection design quality, based on a linear growth DLM with variance learning. Industry guidelines (e.g.…”

Uncertainty Propagation Assessment in Railway-Track Degradation Model Using Bayes Linear Theory

“…The Bayesian approach provides a theoretical inference framework for updating prior beliefs about uncertain quantities once additional information becomes available (if the decision maker can make observations) from the tests and analyses conducted during the development program. An early work on uncertainty assessment using the Bayesian approach has been reported since early 1970 (Randell et al, 2010). Until now, a wide range of extensions has been developed (see review in Lu andMadanat 1994, Zhang andMahadevan 2003), and most of the works were developed under a probabilistic Bayesian framework.…”

Optimal stopping for the predictive maintenance of a structure subject to corrosion

“…The recent literature on optimization of maintenance mainly focuses on comparison of policies [2][3][4]. A different approach is presented here.…”