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

TL;DR

This paper introduces a Bayesian linear variance learning method for analyzing inspection data of large physical systems, improving system state inference and life expectancy forecasts by effectively handling sparse, irregular, and incomplete data.

Contribution

It presents a novel Bayesian linear approach for variance structure learning in large-scale physical system inspections, enhancing inference accuracy over traditional models.

Findings

Materially different remnant life forecasts with variance learning

Effective variance learning from sparse, irregular data

Application to offshore platform pipe-work networks

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 which 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 to an equivalent model neglecting variance learning.