Self-Healing Performance Assessment of Bacterial-Based Healing Concrete Using Machine Learning Approaches - Clarke Lecture 2023

Aug 3, 2026

Upload by Anda

Research investigates and forecasts the self-healing performance of bacterial-based self-healing concrete using machine learning. A data pipeline is built with database preparation, scaling, feature selection, and hyperparameter tuning, then training multiple models to compare predictive accuracy. Results show the GBR model using 22 features achieves the strongest performance, with the highest reported R2 and lowest RMSE among candidates. Feature influence analysis identifies variables with positive and negative effects on healing performance, supporting engineering design and reducing reliance on time- and cost-intensive laboratory testing.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages2

Reading Time5 min

AccessFree to download

Aims and Objectives

Contextual Background

Methods

Data and Processing Pipeline

Feature Selection and Hyperparameter Tuning

Model Training

Results

Prediction Performance

GBR Model with 22 Features

Findings in Context

Conclusions

University of Birmingham Self-Healing Performance Assessment of BacterialBased Healing Concrete Using Machine Learning Approaches Huang, Xu; Kaewunruen, Sakdirat License: Creative Commons: Attribution (CC BY) Citation for published version (Harvard): Huang, X & Kaewunruen, S 2023, 'Self-Healing Performance Assessment of Bacterial-Based Healing Concrete Using Machine Learning Approaches', Clarke Lecture 2023, Birmingham, United Kingdom, 22/06/23-22/06/23 . Link to publication on Research at Birmingham portal General rights Unless a licence is specified above, all rights (including copyright and moral rights) in this document are retained by the authors and/or the copyright holders. The express permission of the copyright holder must be obtained for any use of this material other than for purposes permitted by law. •Users may freely distribute the URL that is used to identify this publication. •Users may download and/or print one copy of the publication from the University of Birmingham research portal for the purpose of private study or non-commercial research. •User may use extracts from the document in line with the concept of ‘fair dealing’ under the Copyright, Designs and Patents Act 1988 (?) •Users may not further distribute the material nor use it for the purposes of commercial gain. Where a licence is displayed above, please note the terms and conditions of the licence govern your use of this document. When citing, please reference the published version. Take down policy While the University of Birmingham exercises care and attention in making items available there are rare occasions when an item has been uploaded in error or has been deemed to be commercially or otherwise sensitive. If you believe that this is the case for this document, [please contact UBIRA@lists.bham.ac.uk](please contact UBIRA@lists.bham.ac.uk) providing details and we will remove access to the work immediately and investigate. Download date: 02. Aug. 2026 Self-Healing Performance Assessment of Bacterial-Based Healing Concrete Using Machine Learning Approaches Xu Huang; Dr Sakdirat Kaewunruen, School of Engineering Aims and Objectives • This research aims to understand and predict the self-healing performance of bacterial-based self-healing concrete by applying machine learning approaches. Contextual Background H2 O O2 CO2 Ca2+ CO32- CaCO3 Supervisor Data Split Output Inputs Algorithms Processing Output Methods Techn ica l Robustness Assurance Database Data Scaling Data Split Data Processing Feature Selection Hyperparameter Tuning VEL Model Models Training Results Prediction performance of the machine learning models Prediction performance of GBR model with 22 features • The GBR model with 22 features presents a relatively superior prediction performance in terms of R2 (0 .950) and RMSE (8 .841%) , compared to the other models. The Findings in Context • The GBR model with 22 features demonstrates the best prediction performance among the other GBR models. • HT, DCI, CA, DN and B have positive effects on HP, while CW, HTM, S and DC show negative effects on HP. FA, HT, CW and C exert a greater influence on HP. Conclusions • The healing performance of bacterial-based healing concrete can be predicted by applying the GBR model with 22 features. • This technique has the potential to assist engineers in the design of bacterial-based healing concrete, saving the time, resources, and costs associated with laboratory tests. Key Publications • Huang, X. , Sresakoolchai, J. , Qin, X. , Ho, Y. F. and Kaewunruen, S. , 2022. Self-Healing Performance Assessment of Bacterial-Based Concrete Using Machine Learning Approaches. Materials, 15 (13), p.4436. • Huang, X. , Wasouf, M. , Sresakoolchai , J. and Kaewunruen, S. , 2021. Prediction of healing performance of autogenous healing concrete using machine learning. Materials, 14 (15), p.4068. • Huang, X. , Ge, J.

FAQs

What is the research goal for bacterial-based self-healing concrete?

To understand and predict the self-healing performance of bacterial-based self-healing concrete using machine learning approaches.

Which machine learning model performs best in the study?

The GBR model using 22 features shows the relatively superior prediction performance, reported with the best R2 and RMSE compared with other models.

How can the method help engineers in practice?

It can assist engineers in designing bacterial-based healing concrete and reduce the time, resources, and costs required for laboratory tests.

Aug 3, 2026

Upload by Anda