Recently, increasing traffic volumes and growing user expectations for expressways in Korea have increased the demand for highperformance bridge deck pavement systems. Rapid-setting concrete has been widely adopted for bridge deck pavement repair because of its high durability and early strength development, enabling rapid traffic reopening. However, its strength development is highly sensitive to the curing conditions, making reliable on-site quality control essential. Conventional quality control based on molded specimen testing has limitations in representing the actual curing conditions of repaired pavements. In this study, a surface-temperature-based maturity model and an IT-based web quality management system were used for bridge deck pavement repair using rapid-setting concrete. Laboratory experiments were conducted under various ambient temperature conditions to establish the relationship between the surface temperature and the compressive strength. A machine-learning-based isolation forest was applied for outlier detection during data preprocessing. Polynomial regression analysis was subsequently performed to develop and validate the maturity prediction models, and a web-based quality management system integrated with an infrared thermographic camera was implemented to estimate the in-place concrete strength in real time. The proposed system provides rapid, objective, and representative quality control compared with conventional specimen-based methods. The developed maturity model and web system have the potential to improve field quality management and support reliable decision-making regarding traffic reopening after bridge deck pavement repair.

This study investigated the influence of nanobubble water generated using carbon dioxide (CO₂) and oxygen (O₂) gases on the physical and chemical performances of cement-based materials, including mortar and concrete. Nanobubbles, defined as gas bubbles with diameters of less than 100 nm, exhibit high surface energy, prolonged stability in aqueous environments, and enhanced chemical reactivity. These properties promote accelerated cement hydration and the development of a refined microstructure. In particular, carbon-dioxide-based nanobubble water facilitates in situ calcium carbonate formation within the cement matrix, contributing to microstructural densification and offering a promising route for permanent carbon dioxide sequestration. Oxygen-based nanobubble water was used as the control because of its stable generation and measurable concentration. Mortar and concrete specimens were prepared using nanobubble water with various gas types and concentrations. The stability of the nanobubble water was evaluated using zeta potential and particle size distribution analyses. The mortar tests included flow and setting time evaluations, whereas the concrete tests consisted of slump, air content, compressive strength at different curing ages (3 and 28 d) and rapid chloride ion penetration resistance measurements. The results showed that carbon-dioxidebased nanobubble water effectively accelerated early hydration, improved early compressive strength, and reduced chloride ion permeability, indicating improved long-term durability. In contrast, oxygen-based nanobubble water resulted in moderate improvements in workability and strength development. These results suggest that the incorporation of nanobubble water, particularly carbon-dioxide-based nanobubble water, into cementitious systems can improve the material performance while supporting carbon-neutral construction through integrated carbon dioxide utilization.

Recent advancements in automotive technology along with sustained investments in road infrastructure and associated transportation facilities have significantly reduced the frequency of road traffic accidents. Nevertheless, human-related factors, particularly driver negligence, continue to account for most traffic incidents. To address accidents caused by driver inattention and drowsiness, this paper proposes a digitalcode- based traffic safety sign system capable of delivering direct and real-time information to drivers through onboard vision-recognition devices, such as vehicle dashcams and black-box systems, which are now widely deployed in modern vehicles. Several candidate digital codes applicable to traffic safety signs were reviewed to develop the proposed system. Among them, the ArUco3 marker was selected owing to its simple geometric structure and superior recognition performance in camera-based detection environments. To experimentally validate the applicability of the proposed system, a prototype mechanism was implemented, in which a vehicle-mounted vision-recognition device detected traffic safety signs embedded with ArUco3 markers and subsequently provided preconfigured warning and guidance information to the driver in real time through audiovisual outputs. Performance evaluation experiments focusing on recognition rate, which is considered the most critical factor governing the practical applicability of the proposed system, were conducted on a real-world test track under actual driving conditions. To ensure a rigorous evaluation, test specimens were fabricated with dimensions equivalent to those of conventional traffic safety signs while varying the fundamental cell size of the embedded ArUco3 markers. Additionally, repeated driving experiments were performed by incrementally increasing the vehicle speed from 30 to 150 km/h. The experimental results demonstrated that the vehiclemounted vision-recognition system successfully detected and recognized the proposed digital traffic safety signs, even under high-speed driving conditions. These findings verify the effectiveness and practical feasibility of the proposed ArUco3-marker-based digital traffic safety sign system. Furthermore, in future heterogeneous traffic environments where autonomous and human-driven vehicles will coexist, the proposed system is expected to provide an additional channel for acquiring road information, thereby enabling more effective responses to unexpected incidents and emergencies.

Predicting pavement deterioration is an essential component of pavement management systems. However, traffic and environmental variables commonly used in prediction models often exhibit high levels of multicollinearity, which may affect the variable selection and model performance. This study investigated the multicollinearity structure of pavement deterioration variables and compared model-specific variable selection characteristics using a genetic algorithm (GA)-based wrapper approach. A dataset consisting of 20,253 observations collected from 30 major arterial roads in Daejeon Metropolitan City was analyzed. Multicollinearity was diagnosed using the variance inflation factor (VIF), and GA-based variable selection was independently applied to multiple linear regression (MLR), random forest (RF), and XGBoost models. Seven of the eight explanatory variables exhibited VIF values greater than 10, indicating substantial multicollinearity. After variable selection, the maximum VIF decreased substantially in both MLR and RF, whereas several highly correlated variables remained in XGBoost. In addition, the selected variable subsets differed across the model types. RF and XGBoost showed improved predictive performances after variable selection, whereas MLR exhibited little performance change despite the reduction in multicollinearity. These findings suggest that the effects of variable selection under multicollinear conditions vary according to the model structure and highlight the importance of model-specific variable selection strategies in pavement deterioration prediction.

Recent climate change has led to a continuous increase in global temperatures and a rise in the frequency and intensity of extreme weather events. Asphalt concrete pavements are particularly sensitive to environmental conditions, and their structural and functional performances are significantly influenced by temperature and moisture. In current pavement design frameworks, such as AASHTO 2002 and the Korea Pavement Research Program, climatic inputs are derived from historical observations and are typically assumed to be stationary, indicating that their statistical properties remain constant over time. However, this assumption may not adequately capture long-term climatic trends, potentially leading to an underestimation of future pavement deterioration. Therefore, this study aimed to evaluate the influence of nonstationary temperature conditions on asphalt pavement performance by incorporating statistically validated climate trends into the MEPDG analysis framework. Hourly temperature data for Gangneung from 1965 to 2024 were collected and analyzed to identify long-term trends and nonstationary characteristics. Stationarity was evaluated using augmented Dickey–Fuller and Kwiatkowski–Phillips–Schmidt–Shin tests across monthly and hourly time series (06:00, 14:00, and 22:00). The results showed that approximately 80.6% of the analyzed time series exhibited nonstationary behavior, indicating statistically significant long-term temperature trends. Based on these findings, future temperature scenarios were constructed considering nonstationary characteristics. A stationary climate scenario was defined as a baseline condition with no climate change, whereas a regression-based scenario was developed to reflect long-term temperature trends. Additionally, two climate change scenarios, SSP2-4.5, and SSP5-8.5, were adopted based on the IPCC projections. These scenarios were converted into hourly climatic inputs compatible with the enhanced integrated climatic model (EICM) within MEPDG. Compared with the stationary condition, the regression-based, SSP2- 4.5, and SSP5-8.5 scenarios showed average temperature increases of approximately 0.83 °C, 1.18 °C, and 2.25 °C, respectively. In particular, for the SSP5-8.5 future temperature scenario, an increase by up to 9.7 °C was observed at 06:00 in April compared with the stationary-based average temperature conditions. When future climate scenarios were applied to the AASHTO 2002 framework, the pavement performance was evaluated under identical structural and traffic conditions, including a 200 mm asphalt concrete surface layer, a 200 mm crushed stone base, a 150 mm subbase, and an A-2-4 subgrade. Pavement distress, including rutting, bottom-up fatigue cracking, top-down fatigue cracking, thermal cracking, and international roughness index (IRI), was predicted and compared across scenarios. The results indicated that rutting and top-down fatigue cracking increased under elevated temperature conditions, whereas bottom-up fatigue cracking, thermal cracking, and IRI exhibited relatively small differences among the scenarios. The total rutting increased from 13.19 mm under the stationary climate condition to 13.71 mm under the regression-based and SSP2-4.5 scenarios, and further to 14.15 mm under the SSP5-8.5 scenario. Similarly, asphalt layer rutting increased from 6.52 mm to 7.43 mm. Top-down fatigue cracking showed the highest sensitivity to temperature changes, increasing from 198.80 m/km under the stationary condition to 209.47 m/km, 208.49 m/km, and 217.76 m/km under the regression-based, SSP2-4.5, and SSP5-8.5 scenarios, respectively. In terms of performance criteria, the design life was maintained at 30 years under stationary climate conditions but decreased to approximately 28 years under the regression-based and SSP2-4.5 scenarios and to approximately 26 years under the SSP5-8.5 scenario because of the earlier exceedance of the top-down cracking criterion. These results indicate that temperature increases associated with nonstationary climatic conditions can accelerate rutting and surface-initiated cracking, leading to a reduction in pavement service life. Therefore, the incorporation of climate change effects into pavement design inputs is necessary for a more realistic performance prediction. These findings demonstrate that conventional pavement design approaches based on stationary climate assumptions may underestimate future pavement deterioration under changing climatic conditions. In particular, nonstationary temperature increases can accelerate key distress mechanisms and potentially reduce pavement service life. Therefore, incorporating nonstationary climate characteristics into pavement design and performance evaluation frameworks is essential for improving the reliability of long-term predictions. Future pavement design, material selection, and maintenance strategies should consider projected climate trends to ensure sustainable infrastructure performance.

This study aimed to evaluate the long-term field performance and microscopic deterioration characteristics of rehabilitation methods applied to alkali–aggregate reaction (AAR)-damaged concrete pavement sections. The effectiveness and limitations of each rehabilitation method were examined by integrating long-term tracking survey results with core-based visual inspection and damage rating index (DRI) analysis. The studied sections were classified into three rehabilitation conditions: a surface-hardener-treated section, a modified stone mastic asphalt (SMA) milling and overlay section, and an untreated jointed concrete pavement (JCP) section. The long-term pavement performance was evaluated using tracking survey data, including the highway pavement condition index (HPCI), international roughness index (IRI), and surface distress. Field cores were collected from each section and examined via visual inspection and DRI analysis. The DRI evaluation considered AAR-related damage features, such as gel-filled voids, cracks in coarse aggregates, reaction rims, cracks in cement paste, and gel-filled cracks. Long-term performance and microscopic deterioration characteristics differed according to the rehabilitation method used. The surface-hardener-treated section maintained a relatively stable HPCI level and showed a smaller increase in surface distress than the untreated JCP section, although its initial IRI remained higher than those of the comparison sections. The modified SMA milling and overlay section exhibited favorable surface functionality and ride quality; however, the DRI results indicated residual AAR-related deterioration in the underlying concrete slab. The untreated JCP section exhibited a lower HPCI and greater crack progression, whereas the sampled cores showed relatively low DRI values, indicating possible spatial variability in internal deterioration. The results indicate that the effectiveness of the rehabilitation methods for AAR-damaged pavement sections cannot be evaluated solely using surface performance indicators. The surface condition, ride quality, crack progression, and internal microscopic deterioration should be interpreted together to assess the rehabilitation effectiveness more rationally. The integrated evaluation framework combining a long-term tracking survey and core-based DRI analysis can support the selection of rehabilitation methods and the establishment of long-term maintenance strategies for AAR-damaged concrete pavements.

This study evaluated the rheological behavior and compressive strength of two cementitious base mortar mixtures with different unit binder contents, water-to-binder ratios, and superplasticizer dosages according to the foam content. Two base mortar mixtures, designated as C500 and C600, with different unit binder contents, water-to-binder ratios, and superplasticizer dosages were prepared. The preformed foams were incorporated at 20%, 40%, and 60% of the base mortar volume. The air content, rheological properties, and compressive strength at 7 and 28d were measured and compared. The air content increased with the foam content. Mixtures containing up to 40% foam exhibited air contents close to the target foam volume immediately after mixing. At a foam content of 60%, the measured air content was lower than the target foam volume, indicating that part of the incorporated foam was not retained during mixing and testing. The flow resistance and compressive strength decreased with increasing foam content. At 28 d, the compressive strengths of the C600-F60 and C500-F60 mixtures were 5.6 MPa and 6.4 MPa, respectively. The 40% foam mixtures exhibited relatively high air retention immediately after mixing and reduced rheological parameters; however, their 28-d compressive strengths exceeded the 8.3 MPa limit specified for controlled low-strength materials (CLSM). The 60% foam mixture satisfied the CLSM strength criterion and could be applicable as a low-strength filling material for sewer pipeline voids. As this study compared two base mortar mixtures with different unit binder contents, water-to-binder ratios, and superplasticizer dosages, the observed differences should not be interpreted as an independent effect of the unit binder content.

The increasing number of waste solar panels has created a need for effective recycling technologies. This study investigated the feasibility of using waste-tempered glass from solar panels as a fine aggregate in base-course asphalt mixtures. Waste glass aggregate was produced by crushing and screening waste solar panels, and 20% of the fine aggregate was replaced with waste glass aggregate. The performances of the asphalt mixtures were evaluated using IDEAL-CT, IDEAL-Rut, dynamic modulus, flow number, and four-point bending tests. The pavement service life and life-cycle cost were also analyzed using the TxME and RealCost programs.The laboratory results showed that the asphalt mixture containing waste glass aggregate exhibited improved cracking resistance, rutting resistance, fatigue resistance, and hightemperature performance compared with the conventional asphalt mixture. In addition, the predicted pavement service life was extended, and the life-cycle cost was reduced by 31.6%. These results indicate that waste tempered glass from solar panels can be effectively recycled as fine aggregates in asphalt mixtures while improving pavement performance and economic efficiency.

This study was conducted to develop a machine learning model that classifies hazardous winter road surface conditions, thereby supporting snow removal decision-making in place of the visual inspection and operator experience currently relied upon in practice. Road surface imagery and surface temperature were collected at 10-minute intervals from four field sites in Seoul during the 2023-2024 winter season, and were merged with meteorological data obtained from the nearest weather stations. Road surface conditions were labeled into two classes according to accident risk: "Ice or Snow" and "Normal or Wet." The Synthetic Minority Over-sampling Technique (SMOTE) was then applied to the training set only, leaving the test set unaltered. A random forest classifier was trained, and twelve input variable cases were compared to determine whether excluding highly correlated variables improves predictive performance. Feature importance was assessed using both Gini importance and permutation importance to verify the robustness of the results. Excluding highly correlated variables generally degraded predictive performance rather than improving it, indicating that air temperature, surface temperature, dewpoint, and relative humidity each retain unique information that cannot be fully explained by the others. The best-performing input set excluded only wind speed, achieving a recall of 0.96, an F1-score of 0.90, and a precision of 0.86 for the hazardous class. Wind speed exhibits strong local variability, and values measured at weather stations were therefore considered inadequate for representing field conditions. The two feature importance measures produced largely consistent rankings, with surface temperature, relative humidity, and time elapsed since the end of precipitation ranking highest. Notably, surface temperature ranked first in permutation importance despite its high correlation with air temperature and dewpoint, indicating that it carries information that cannot be substituted by other temperature-related variables. A random forest model was developed to classify hazardous winter road surface conditions from meteorological data and surface temperature. Surface temperature and relative humidity were identified as the dominant variables. Since the model relies on variables observed at a single point in time rather than timeseries structures, it can be extended to predict future road surface conditions.

Autonomous Environmental Service Vehicles (AESVs) require safety verification that is distinct from passenger AVs because of their dual-task nature. This study proposes a dual-perspective ODD risk assessment framework using three months of field data (Oct–Dec 2025) from an AESV deployed in Jeju City, South Korea. The PCA-weighted Enhanced CII quantifies the internal mechanical interference, whereas entropy-weighted risk scores combined with k-means clustering identify external hazard zones. The CII declined from 1.25 to 1.12, confirming progressive adaptation of compensatory control, whereas spatial mapping revealed persistent hotspots at intersections, sharp curves, and commercial districts, indicating that ODD boundaries are governed by road geometry and land use.

This study analyzes changes in the road-user conflict characteristics of personal mobility (PM) crashes in Seoul before and after safety regulations were strengthened on May 13, 2021. 2,399 police-reported PM crashes from 2017 to 2024 were reconstructed at the crash level. Crash types were classified as PM-pedestrian conflicts, PM-vehicle conflicts, or single-PM crashes. The primary comparison used two-year pre- and post-policy windows, and an additional one-year comparison was conducted to examine short-term changes. Monthly segmented count models and multinomial logistic regression models were applied as supplementary analyses. In the two-year comparison, crash counts increased after the regulation was implemented; however, the crash type, severity, and time-period composition did not change significantly. In the one-year comparison, crash counts decreased and the crash type composition changed significantly. The multinomial logistic regression results also showed a short-term increase in the relative share of PM-vehicle crashes compared with PM-pedestrian crashes; however, this change was not sustained in the two-year comparison. Changes in the conflict characteristics of PM crashes after safety regulations were partially observed in the short term, but they cannot be interpreted as persistent medium-term changes. Because exposure and enforcement data were unavailable, the findings should be interpreted as observed changes in police-reported crash characteristics rather than as the causal effects of regulation.

Autonomous Demand-Responsive Transit (Autonomous DRT) is a public transportation service that integrates autonomous driving technology with DRT. The success of Autonomous DRT depends on operational performance as well as public acceptance. Therefore, both expert and public perspectives need to be considered in service planning and evaluation. This study compares expert and public perceptions of the evaluation criteria for Autonomous DRT services using a common evaluation framework. Expert priorities were obtained from the analytic hierarchy process (AHP) results of a previous study involving 32 experts, whereas public priorities were analyzed using survey data from 300 residents of Namyang-eup, Hwaseong-si. Pairwise comparison questions were used as evaluation items to reduce the response burden on the public respondents, and ranking questions were used as evaluation indicators. The ranking data were converted into AHP-type importance values. Analyses using multiple weighting methods and nonparametric bootstrap procedures were conducted to examine the stability of the estimated priorities and their sensitivity to the weighting transformations. The results show that experts and the public evaluate Autonomous DRT services from different perspectives and assign different priorities to the evaluation criteria. The experts placed relatively greater importance on system-level performance factors, such as operational efficiency, service reliability, and demand responsiveness. In contrast, public respondents tended to prioritize factors directly related to the actual user experience, including access distance to boarding points, driving speed stability, and the number of vehicles in operation. These findings suggest that the planning and evaluation of Autonomous DRT services should consider system-level indicators of operational efficiency along with user-perceived accessibility, safety, and service availability. This study highlights the need to develop Autonomous DRT service designs and evaluation frameworks that balance system efficiency with user acceptance by incorporating both expert-oriented operational performance indicators and public-oriented perceived service quality indicators.

Bicycle crashes in urban areas are frequently concentrated at intersections where motor vehicles, pedestrians, and cyclists share the same roadway space. Therefore, there is an increasing need for a hotspot identification methodology that considers both the spatial characteristics of crash locations and crash severity. However, conventional hotspot identification methods primarily rely on crash frequencies aggregated by administrative districts or intersections, making it difficult to identify the specific hazardous locations within intersections where crashes actually occur. This study proposes a methodology for identifying urban bicycle crash hotspots by incorporating the spatial characteristics of bicycle crashes. Bicycle crash data collected in Gwangsan-gu, Gwangju Metropolitan City, from 2020 to 2024 were analyzed using the Traffic Accident Analysis System (TAAS) Geographic Information System (GIS). The distance between each crash location and the nearest intersection corner was calculated to represent the spatial influence of intersections. Crash severity was quantified using Equivalent Property Damage Only (EPDO) weights, while distance-based weights were established from the cumulative distribution of crash locations. A bicycle crash risk index was developed by integrating crash severity and distance weights, enabling differentiated risk evaluation even for locations with identical crash frequencies. Furthermore, sensitivity analysis was conducted by varying the distance-weighting scheme to verify the robustness and applicability of the proposed methodology. The results demonstrated that the proposed approach identified hazardous locations more precisely than conventional intersection-based methods and maintained stable ranking results. The proposed methodology can improve the objectivity and reliability of bicycle hotspot identification and provide practical support for prioritizing bicycle safety improvement projects, planning intersection safety measures, establishing transportation safety policies, and allocating limited public resources more efficiently.