Techno Press
Tp_Editing System.E (TES.E)
Login Search
You logged in as...

gae
 
CONTENTS
Volume 47, Number 1, October10 2026 (Special Issue)
 

Abstract
As the utilization of underground space has increased due to the saturation of ground surface space, the construction of tunnels using shield TBM (Tunnel Boring Machine) has expanded. Inspection and replacement of TBM cutters under high water pressure conditions suffer from significantly reduced workability owing to excessive groundwater inflow. To address this issue, a previous study [7] proposed NFGM (New Face Grouting Method) in which grout is injected into the ground ahead of the tunnel face from the TBM chamber to stabilize the face and reduce water inflow. In this study, a numerical model was developed to quantitatively evaluate the applicability of the proposed NFGM and to investigate the grout penetration characteristics under varying ground conditions. A coupled VOF–DEM numerical approach, combining the Volume of Fluid (VOF) model and the Distinct Element Method (DEM) implemented in ANSYS FLUENT, was employed to simulate the flow behavior of grout and the movement of soil particles within water-saturated sandy ground. The porosity, particle diameter, and relative density of the ground were considered as key parameters, and the grout penetration distance was analyzed with respect to variations in injection pressure. In addition, an analytical equation for estimating grout penetration distance suitable for practical applications was established based on a review of existing theories, and a comparative analysis between the numerical results and the analytical predictions was conducted to identify the boundary between permeation grouting and fracture grouting.

Key Words
grouting; FLUENT; fluid flow; fracture grouting; permeation grouting; tunnel boring machine

Address
Pill-Bae Hwang, Terry Moon, Seok-Won Lee: Department of Civil and Environmental Engineering, Konkuk University,
120 Neungdong-ro, Gwangjin-gu, Seoul 05029, Republic of Korea

Abstract
The infiltration behavior of bentonite slurry substantially affects excavation face stability in slurry shield tunnel boring machine (TBM) excavation. To investigate slurry infiltration into saturated sand, this study conducted a series of infiltration column tests under varying applied pressures and slurry concentrations. Measurements of discharged water enabled the identification of two distinct infiltration stages: an initial slurry invasion phase followed by a filter cake formation phase. The onset of the filter cake formation phase was evaluated using both blocking filtration laws and the Peclet number–based method, and the discrepancy between the results obtained from these two approaches remained negligible across all test conditions. Moreover, it was observed that filter cake formation occurred more rapidly at higher applied pressures and slurry concentrations. During the slurry invasion stage, the penetration distance increased with applied pressure but decreased with increasing slurry concentration. In addition, the filtration equation was employed to estimate the hydraulic conductivities of the slurry-penetrated zone and the filter cake using revised experimental data. The estimated hydraulic conductivities were on the order of 10-7 - 10-6 m/s for the slurry-penetrated zone and approximately 10-10 m/s for the filter cake, both of which are significantly lower than that of the sand bed (2.5 x 10-4 m/s). Notably, the hydraulic conductivities of both the slurry-penetrated zone and the filter cake decreased as slurry concentration and applied pressure increased.

Key Words
filter cake formation; hydraulic conductivity; infiltration tests; slurry invasion; slurry shield tunneling

Address
Yuemyung Yoon: Department of Geotechnical Engineering Research, Korea Institute of Civil Engineering and Building
Technology, 283, Goyangdae-ro, Ilsanseo-gu, Goyang-si, Gyeonggi-do, Republic of Korea
Hyunrae Kim: Infrastructure Division, Hyundai Engineering and Construction, 75, Yulgok-ro,
Jongno-gu, Seoul, Republic of Korea
Joohyeon Seo: Department of Civil, Environmental and Architectural Engineering, Korea University, 145, Anam-ro,
Seongbuk-gu, Seoul, Republic of Korea
Hangseok Choi: School of Civil, Environmental and Architectural Engineering, Korea University, 145, Anam-ro,
Seongbuk-gu, Seoul, Republic of Korea

Abstract
Evacuation efficiency in deep-underground stations is highly sensitive to disaster-induced congestion and dynamic pedestrian behavior. This study introduces a real-time evacuation simulation framework that evaluates and adapts optimal evacuation routes based on the progression of fire and the spatial distribution of evacuees. Instead of relying on fixed assumptions, this model integrates congestion-sensitive travel time recalculation into the pathfinding process. A virtual six-level underground station is constructed as a node-edge graph, and five fire scenarios are implemented by altering node accessibility. The simulation demonstrates that blockage at highly congested zones causes significantly longer evacuation durations, highlighting the importance of hazard-aware route reallocation in underground station design.

Key Words
congestion; deep underground station; Dijkstra's algorithm; evacuation simulation; fire scenario; network analysis

Address
Mintaek Yoo, Hyunseok Kim: Department of Civil & Environmental Eng., Gachon University, 1342 Seongnam-daero,
Seongnam-si 13120, Republic of Korea
Changwon Kwak: Department of Civil and Environmental Engineering, Inha Technical College, 100, Inha-ro, Incheon, Korea
Seongwon Hong: Department of Safety Engineering, Korea National University of Transportation, 50 Daehak-ro,
Chungju-si, ChungBuk 27469, Republic of Korea

Abstract
This research evaluates approximate fundamental period coefficients for pile-supported structures, with a specific focus on the impacts of Soil-Pile-Structure Interaction (SPSI). Existing seismic design codes typically utilize empirical period equations derived from shallow foundation data assuming rigid base conditions, which frequently fail to account for the period elongation phenomena observed in pile-supported systems on soft ground. To bridge this gap, this study introduces new approximate period formulas tailored for pile-supported buildings, developed through a combination of 1 g shaking table tests and three-dimensional finite difference analyses (FLAC 3D). The proposed equations incorporate coefficients (Ct and x) that explicitly represent the flexibility of the soil-pile system. Furthermore, the study extends its scope to examine seismic design implications by integrating the proposed period equations with standard design response spectra (e.g., ASCE7-16 and KDS 41 17 00). The findings indicate that accounting for SPSI results in a marked elongation of the fundamental period, which in turn leads to a reduction in spectral design acceleration (Sa) relative to the rigid-base assumption. This reduction highlights that ignoring SPSI may result in excessively conservative seismic force estimates, whereas the proposed approach provides a more rational and cost-effective basis for the preliminary seismic design of pile-supported structures.

Key Words
liquefaction; pile-structure connection; seismic design; shaking-table test; soil-pile-structure interaction

Address
Sumin Song: Department of Coastal Construction Engineering, Kunsan National University, Kunsan, Republic of Korea
Junyoung Ko: Department of Civil Engineering, Chungnam National University, Daejeon, Republic of Korea

Abstract
Groundwater behaviour induced by tunnel excavation can be interpreted as a hydraulic convergence-confinement problem, in which inflow-pressure interact depending on ground and support conditions at the excavation boundary. The recently proposed Hydraulic Convergence-Confinement Method (H-CCM) provides a theoretical framework for describing hydraulic equilibrium behaviour through the inflow-pressure relationship; however, its applicability is limited by simplified hydraulic boundary conditions. In this study, the H-CCM is extended by introducing hydraulic confinement control elements that explicitly account for practical inflow control measures, including grouting, pinhole drainage, and excavation support, as well as their combined application. In particular, grouting previously treated as a ground-related inflow reduction factor is redefined as a hydraulic support element, enabling a unified description of hydraulic support behaviour under combined grouting-lining and pinhole support conditions. The proposed framework is examined through theoretical parametric analyses and numerical simulations, demonstrating its capability to systematically capture hydraulic equilibrium behaviour under various ground and support configurations. By allowing arbitrary hydraulic boundary conditions between fully drained and undrained extremes, the extended H-CCM enables identification of hydraulic equilibrium points that satisfy prescribed allowable inflow rates. The proposed approach provides a rational basis for the preliminary design of tunnel inflow control and groundwater drawdown measures.

Key Words
combined hydraulic support systems; hydraulic confinement control; hydraulic equilibrium point

Address
Seung-Hyun Kim: Engineering division, Geo Expert Group CO., Formerly konkuk University, Gyeonggi-do, Korea
Jae-Ho Jeong: Engineering division, Geo Expert Group CO., Gyeonggi-do, Korea
Jong-Ho Shin: Marugeo Research Institute, Gyeonggi-do, Korea, Formerly Konkuk University, Korea


Abstract
As urban underground development expands, mechanical excavation methods are increasingly adopted to reduce vibration, noise, and excavation-induced damage. However, tunnel support patterns for road-header excavation are still commonly derived from drill-and-blast practice, which can lead to conservative support quantities. This study evaluates the feasibility of optimizing road-header support by considering excavation-methoddependent excavation damaged zones (EDZs) and the RMR14 classification system. Two-dimensional finiteelement analyses using PLAXIS 2D were performed with an EDZ thickness of 2.0 m for drill-and-blast excavation and 0.5 m for road-header excavation. Tunnel displacement, rock-bolt axial force, shotcrete stress, and support safety factors were compared under progressively reduced support cases. The road-header model produced substantially smaller deformation, with unsupported crown displacement approximately 44% of that for drill-and-blast excavation and an average supported crown displacement of approximately 51%. Normalized rock-bolt force and shotcrete bending-compressive stress remained lower, averaging approximately 0.84 and 0.74 of the drill-and-blast values, respectively. RMR14-based analyses further indicated that support reduction is feasible in Classes II-III, whereas Class IV requires a minimum shotcrete thickness of 100 mm under the analyzed conditions. The results demonstrate that excavation-method effects can provide a rational basis for support optimization, while site-specific geological conditions and construction-stage monitoring remain essential for practical application.

Key Words
excavation damaged zone; numerical analysis; rmr14; road-header; support optimization

Address
Hyoung-Seok Oh, Joon-Shik Moon: Department of Civil Engineering, Kyungpook National University, Daegu 41566, Republic of Korea
Dae-Hyun Kang: Geotechnical Engineering and Tunnel Dept., DASAN Consultants, Seoul 05800, Republic of Korea

Abstract
In recent years, slurry TBM (Tunnel Boring Machine) method has been used successfully in urban areas, under river, and undersea tunnels. However, TBM operations can pose risks such as ground settlement or collapse, often resulting from over-excavation. Therefore, the excavation should be monitored during tunneling to prevent the over- excavation. Although slurry TBM can form a bentonite filter cake that helps prevent over-excavation, excavation management remains challenging because excavated materials circulate in the closed-circuit. Therefore, sensor data such as flow meter and density meter are used to calculate the over-excavated materials amount by theoretical method. There are conventional methods using in-situ as well as dry density, but they cannot account for the scenario where excess water infiltrates from the surrounding ground. This study modified a method to calculate both the in-situ over-excavation and the water inflow separately. The suggested method was applied in a slurry TBM site, where high water pressure and mixed ground condition exist, to evaluate the over excavation more accurately. The results showed a good agreement with actual measurement from the site. Also, the simplified equation was suggested to easily detect over-excavation.

Key Words
filter cake test; modified method; over excavation; slurry; TBM; water inflow

Address
Jaehoon Jung: R&D center, Hyundai Engineering & Construction, 75, Yulgok ro, Jongno-gu, Seoul, Republic of Korea;
Environmental & Architectural Engineering, Korea University, 145, Anam ro, Seongbuk-gu,
Seoul, Republic of Korea
Jaewon Lee,Young Jin Shin: R&D center, Hyundai Engineering & Construction, 75, Yulgok ro, Jongno-gu, Seoul, Republic of Korea
Kyuhyeong Lim: Environmental & Architectural Engineering, Korea University, 145, Anam ro, Seongbuk-gu,
Seoul, Republic of Korea

Abstract
Obtaining reliable interpolation of soil properties in underground spaces is challenging due to the limited number of site investigations and inherent spatial variability of soils. This study investigates the impact of site investigation depths and patterns on the performance of machine learning-based three-dimensional (3D) geotechnical interpolation. The concept of geotechnical distance field was adopted for 3D interpolation and extreme gradient boosting (XGB) was selected as the high-performing machine learning algorithm to train dataset including geotechnical distance and cone tip pressure. It was found that reliable interpolation can be achieved by ten site investigations to cover 25 m x 25 m using 3D GDF-XGB model. In addition, the variability of the length of site investigation can be quantitatively assessed using the framework shown in this study, in which the intermediate level of performance can be still achieved with high variability of the length of site investigations. The framework to determine the location of optimal site investigations and determining the optimal interval between site investigations for irregular geological surface shown in this study can apply to the construction projects that need reliable 3D interpolation of geotechnical properties.

Key Words
geotechnical distance field; machine learning; site investigation; spatial interpolation; underground space

Address
Tran-Gia-Khiem Nguyen, Jongmuk Won: Department of Civil, Environmental and Architectural Engineering, Korea University, 145 Anam-ro,
Seongbuk-gu, Seoul 02841, Republic of Korea

Abstract
To investigate ground anomalies behind a cementitious wall, traditional invasive techniques such as coring and drilling have been adopted, which may significantly disturb and destabilize the target ground or construction materials. In this study, the ground penetrating radar (GPR) was employed to non-invasively characterize subsurface materials behind a cementitious wall. To simulate the cementitious walls with different thicknesses, two cuboid cementitious specimens were prepared. In addition, air, dried sand, saturated sand, and water were filled in the space between the two specimens to simulate possible subsurface materials behind a cementitious wall. GPR surveys were performed on both surfaces of the cementitious walls for cases with different filling materials between the walls, and the reflected electromagnetic (EM) waves were collected in time domain. Under all experimental conditions, the electromagnetic waves reflected from the specimen surfaces and the interfaces between specimens and filling materials were clearly identified in the GPR signals. The depth of the filling materials evaluated based on the travel time of the reflected EM waves agrees with the thickness of the walls. In addition, the amplitudes of the reflected EM waves, which varies according to the experimental conditions such as thickness of the walls and the type of filling materials, are quantitatively analyzed in relation to the experimental conditions. The GPR signal analysis method proposed in this study can be effectively applied to non-invasively identify ground anomalies behind a cementitious wall.

Key Words
attenuation characteristics; electromagnetic wave; ground anomaly characterization; ground penetrating radar; reflection characteristics

Address
Heechang Hwang, Eun Sang Lee, Won-Taek Hong: Department of Civil & Environmental Engineering, Gachon University,
1342, Seongnam-daero, Sujeong-gu, Seongnam-si, Gyeonggi-do, 13120, Republic of Korea
Young-Sam Kwon: Smart Construction Promotion Center, Korea Institute of Civil Engineering and Building Technology
283, Goyang-daero, Ilsanseo-gu, Goyang-si, Gyeonggi-do, 10223, Republic of Korea

Abstract
Optimizing charging density in tunnel blasting is critical for excavation efficiency and stability. However, traditional approaches apply uniform charging across tunnel rounds, ignoring the distinct requirements of different blast zones. While previous studies have acknowledged these zone-specific needs, a comprehensive machine learning framework for predicting charging density in each zone has been lacking. This study addresses that gap by developing a zone-specific framework to predict the optimal charging density for the cut, stoping, lift, and contour zones. Using data from 334 tunnel blast rounds from 23 Korean sites representing diverse geological conditions, we evaluated various machine learning algorithms including linear models, tree-based ensembles, and gradient boosting methods. The exploratory data analysis revealed that charging density distributions, sensitivity to Rock Grade, ranges of blast design parameters, and correlation structures differ substantially across zones, providing strong justification for adopting a zone-specific modeling strategy. Our evaluation found that tree-based ensemble models achieved strong predictive performance for all zones, with the contour and stoping zones showing the highest accuracy due to their standardized design practices. Furthermore, SHAP analysis revealed that each zone is governed by distinct parameters, with rock mass properties dominating cut zone predictions, while blast design parameters are most significant for stoping and lift zones, and design control-related variables are most influential for the contour zone. The proposed framework provides engineers with a practical, data-driven tool to improve fragmentation efficiency, reduce overbreak, and enhance both safety and cost-efficiency in tunnel excavation.

Key Words
blast design optimization; charging density; machine learning; SHAP analysis; tunnel blasting; zone-specific prediction

Address
Jung In Kwon, Tae Young Ko: Department of Energy and Resources Engineering, Kangwon National University,
1 Gangwondaehak-gil, Chuncheon 24341, Republic of Korea
Ho Seong Lee: Jiyun E&C, 25 Beobwon-ro 11-gil, Songpa-gu, Seoul 05836, Republic of Korea

Abstract
In urban twin tunnel construction, ground behaviour can vary significantly depending on the excavation method, and there is a growing demand for technologies capable of quantitatively evaluating and predicting these differences. This study aims to compare and analyze the soil shear deformation characteristics associated with sequential and simultaneous twin tunneling in soft ground conditions and to examine the applicability of a machine learning-based shear deformation prediction model trained using numerical analysis results. Two-dimensional finite element analyses were performed based on laboratory model tests, and 1,296 numerical analysis scenarios were generated using a Python-PLAXIS integrated automation framework. Spatial shear deformation data were extracted from the analysis results to construct a training database, and the predictive performance of the Random Forest, SVM, XGBoost, and LightGBM algorithms was compared. The results show that the XGBoost model achieved the best predictive performance, with further improvements in accuracy and stability through Bayesian-based hyperparameter optimization. Visualization of shear deformation demonstrates that sequential tunneling induces concentrated deformation around the piles adjacent to the outer wall, whereas simultaneous tunneling exhibits a more symmetrical and moderate distribution consistent with the numerical analysis results, thereby validating the potential of a data-driven approach for the spatial prediction and visualization of shear deformation under twin-tunnel construction conditions.

Key Words
machine learning; numerical analysis; prediction model; shear deformation; twin tunnel

Address
Subin Kim,Hyung-Joon Seo, Yong-Joo Lee: Department of Civil Engineering, Seoul National University of Science and Technology,
232 Gongneung-ro, Nowon-gu, Seoul 01811, Republic of Korea
Dong-Wook Oh: Research and Development Institute, Hanul D&B Co., Ltd., 178, Digital-ro, Geumcheon-gu,
Seoul 08513, Republic of Korea

Abstract
For subways, ground settlement can develop above the tunnel during both construction and the maintenance period under the combined effects of construction activities and surrounding environmental conditions. Because the locations where settlement may appear along the subway corridor are difficult to anticipate in advance, a wide-area forensic monitoring approach is needed; however, conventional measurement techniques have practical limitations in observing an entire corridor simultaneously. To address this, multi-year Interferometric Synthetic Aperture Radar (InSAR) observations were used to conduct wide-area forensic monitoring of ground change above the Bundang Line of Seoul Metro, South Korea. This study analyzed a 16 km section of the Bundang Line (Suseo Station–Cheongnyangni Station) using InSAR data acquired from February 2018 to January 2025. The Small Baseline Subset (SBAS)-InSAR method was adopted to mitigate decorrelation effects associated with surrounding buildings and vegetation in the monitored area. Based on the InSAR results, settlement was identified around a ventilation shaft, and the study further determined SBAS processing parameters suitable for detecting surface settlement along the Bundang Line corridor.

Key Words
InSAR time-series Analysis; SBAS-InSAR; Seoul Metro; surface settlement monitoring; ventilation shaft

Address
Taeyong Park, Sunil Cha,Hyungjoon Seo: 1Department of Civil Engineering, Seoul National University of Science and Technology, Seoul, South Korea, 01811
Lang Fu: Department of Civil Engineering and Industrial Design, University of Liverpool, Liverpool, UK, L69 7WW
Jinho Choo: Department of Infrastructure Safety, Korea Authority of Land & Infrastructure Safety,
Gangwon, 201, South Korea
Jong-Sub Lee: School of Civil, Environmental and Architectural Engineering, Korea University, Seoul, South Korea, 02841

Techno-Press: Publishers of international journals and conference proceedings.       Copyright © 2026 Techno-Press ALL RIGHTS RESERVED.
P.O. Box 33, Yuseong, Daejeon 34186 Korea.
General Inquiries: info@techno-press.com / Journal Administration: admin@techno-press.com