Machine learning techniques for structural health monitoring of heritage buildings: A state-of-the-art review and case studies

This paper performed a systematic review of the various machine learning (ML) techniques applied to assess the health condition of heritage buildings. More robust predictive models can be obtained through the effective utilisation of test data gathered from the laboratory or field combined with ML. These models can be used for several predictive applications such as predicting the compressive strength of masonry or repair mortars, possible damage scenarios in heritage buildings, seismic vulnerability assessment, determination of the mechanical properties of materials, and superficial damages on the surface of the monument due to weathering effects, material loss, efflorescence, seepage, algae growth, and moss deposition. The case studies of several interesting and diverse applications of ML, such as its use in strengthening applications, are also discussed. Indirect factors that must be determined to quantify the extent of damage such as moisture content, cracks, and amount of deposited dust in heritage buildings, are also discussed. Furthermore, future research directions and challenges in applying the ML techniques in the case of heritage buildings are highlighted. (C) 2020 Elsevier Masson SAS. All rights reserved.

Automated summary

This paper systematically reviews the application of machine learning techniques in assessing the health condition of heritage buildings, highlighting the importance of combining test data with ML to obtain more robust predictive models. Various applications of ML in predicting compressive strength, damage scenarios, seismic vulnerability, and other aspects of heritage buildings are discussed, along with the consideration of indirect factors for quantifying damage extent. Future research directions and challenges in using ML techniques for heritage buildings are also emphasized.