# Hierarchical Bayesian Modelling for Knowledge Transfer Across Engineering Fleets via Multitask Learning

arXiv: https://arxiv.org/abs/2204.12404 · HTML: https://mlsanity.com/paper/2204.12404/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- Builds on 8 of its 77 references that are in the map.
- First posted on arXiv 26 Apr 2022.

## Builds on (most cited first, 8 of 8)

- [Challenges in Deploying Machine Learning: a Survey of Case Studies](https://mlsanity.com/paper/2011.09926/) (2020)
- [Correlation Alignment for Unsupervised Domain Adaptation](https://mlsanity.com/paper/1612.01939/) (2016)
- [Vector-Valued Control Variates](https://mlsanity.com/paper/2109.08944/) (2021)
- [Domain Adaptive Transfer Learning for Fault Diagnosis](https://mlsanity.com/paper/1905.06004/) (2019)
- [Domain Adaptation for One-Class Classification: Monitoring the Health of Critical Systems Under Limited Information](https://mlsanity.com/paper/1907.09204/) (2019)
- [Bayesian Modelling of Multivalued Power Curves from an Operational Wind Farm](https://mlsanity.com/paper/2111.15496/) (2021)
- [Bayesian Assessments of Aeroengine Performance with Transfer Learning](https://mlsanity.com/paper/2011.14698/) (2020)
- [On an application of graph neural networks in population based SHM](https://mlsanity.com/paper/2203.01646/) (2022)

## Built on by (most cited first, 11 of 11)

- [On the hierarchical Bayesian modelling of frequency response functions](https://mlsanity.com/paper/2307.06263/) (2023)
- [Quantifying the value of information transfer in population-based SHM](https://mlsanity.com/paper/2311.03083/) (2023)
- [When is an SHM problem a Multi-Task-Learning problem?](https://mlsanity.com/paper/2305.09425/) (2023)
- [A decision framework for selecting information-transfer strategies in population-based SHM](https://mlsanity.com/paper/2307.06978/) (2023)
- [Multitask learning for improved scour detection: A dynamic wave tank study](https://mlsanity.com/paper/2408.16527/) (2024)
- [Active learning for regression in engineering populations: A risk-informed approach](https://mlsanity.com/paper/2409.04328/) (2024)
- [A Meta-Learning Approach to Population-Based Modelling of Structures](https://mlsanity.com/paper/2302.07980/) (2023)
- [Sharing Information Between Machine Tools to Improve Surface Finish Forecasting](https://mlsanity.com/paper/2310.05807/) (2023)
- [Towards Multilevel Modelling of Train Passing Events on the Staffordshire Bridge](https://mlsanity.com/paper/2403.17820/) (2024)
- [Inter-turbine Modelling of Wind-Farm Power using Multi-task Learning](https://mlsanity.com/paper/2502.14527/) (2025)
- [Formal Bayesian Transfer Learning via the Total Risk Prior](https://mlsanity.com/paper/2507.23768/) (2025)

## Abstract (arXiv)

A population-level analysis is proposed to address data sparsity when building predictive models for engineering infrastructure. Utilising an interpretable hierarchical Bayesian approach and operational fleet data, domain expertise is naturally encoded (and appropriately shared) between different sub-groups, representing (i) use-type, (ii) component, or (iii) operating condition. Specifically, domain expertise is exploited to constrain the model via assumptions (and prior distributions) allowing the methodology to automatically share information between similar assets, improving the survival analysis of a truck fleet and power prediction in a wind farm. In each asset management example, a set of correlated functions is learnt over the fleet, in a combined inference, to learn a population model. Parameter estimation is improved when sub-fleets share correlated information at different levels of the hierarchy. In turn, groups with incomplete data automatically borrow statistical strength from those that are data-rich. The statistical correlations enable knowledge transfer via Bayesian transfer learning, and the correlations can be inspected to inform which assets share information for which effect (i.e. parameter). Both case studies demonstrate the wide applicability to practical infrastructure monitoring, since the approach is naturally adapted between interpretable fleet models of different in situ examples.