Tunnel overbreak and underbreak prediction using measurement-while-drilling and reconstructed blasthole geometry

Overbreak and underbreak along the excavation profile were predicted at the peripheral-hole scale in drill-and-blast tunnelling. A data set of 537 peripheral holes from 12 consecutive blasting cycles in a tunnel was compiled, including measurement-while-drilling logs, charge-structure records, reconstructed blasthole geometry and post-blast point-cloud profiles. For each hole, 50 depth points were sampled from 1 m below the hole top to the hole bottom. Geometric and charging descriptors derived from blasting theory were computed along depth, including local hole spacing, burden to a reconstructed free surface, concentration factors and charge presence. Summary statistics (mean, quartiles and interquartile range) were computed to represent each hole. A gradient-boosting decision-tree model was trained to predict the blasthole-deviation-corrected overbreak/underbreak value at sampling points and evaluated using cross-validation. The model achieved a root mean square error of 0.137 m, a mean absolute error of 0.096 m and a coefficient of determination of 0.773. Interpretation using Shapley additive explanations quantified the contributions of geometric, charging and drilling-response features.

Authors

Institutions

Publication Details

Journal
Proceedings of the Institution of Civil Engineers - Geotechnical Engineering
Published
2026-10-05
DOI
https://doi.org/10.1680/jgeen.26.00069
Primary Topic
Blasting Impact and Analysis
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Tunnel overbreak and underbreak prediction using measurement-while-drilling and reconstructed blasthole geometry

Lijie Yu, Wanzhi Zhang, Shikui Jia, Bangshu Xu et al.
Proceedings of the Institution of Civil Engineers - Geotechnical Engineering
Blasting Impact and Analysis
article

Tunnel overbreak and underbreak prediction using measurement-while-drilling and reconstructed blasthole geometry

Lijie Yu, Wanzhi Zhang, Shikui Jia, Bangshu Xu, Wanxin Zhu
article en

Abstract

Overbreak and underbreak along the excavation profile were predicted at the peripheral-hole scale in drill-and-blast tunnelling. A data set of 537 peripheral holes from 12 consecutive blasting cycles in a tunnel was compiled, including measurement-while-drilling logs, charge-structure records, reconstructed blasthole geometry and post-blast point-cloud profiles. For each hole, 50 depth points were sampled from 1 m below the hole top to the hole bottom. Geometric and charging descriptors derived from blasting theory were computed along depth, including local hole spacing, burden to a reconstructed free surface, concentration factors and charge presence. Summary statistics (mean, quartiles and interquartile range) were computed to represent each hole. A gradient-boosting decision-tree model was trained to predict the blasthole-deviation-corrected overbreak/underbreak value at sampling points and evaluated using cross-validation. The model achieved a root mean square error of 0.137 m, a mean absolute error of 0.096 m and a coefficient of determination of 0.773. Interpretation using Shapley additive explanations quantified the contributions of geometric, charging and drilling-response features.

Proceedings of the Institution of Civil Engineers - Geotechnical Engineering
Xihua University (CN), Qilu University of Technology (CN), Shandong University (CN), HBIS (China) (CN), Shandong Transportation Research Institute (CN)
Openalex Percentile: Top 15%
Blasting Impact and Analysis
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.