Civil Engineering Journal
Not a member yet
2007 research outputs found
Sort by
A Comparative Study of a Series of Supervised Learning Models for Motorcycle Crash Injury Severity Prediction
Motorcycle crashes pose a major public health challenge in Thailand, where motorcyclists account for most traffic fatalities. This study aims to evaluate and compare the predictive performance of four supervised learning models—Decision Tree (DT), K-Nearest Neighbor (KNN), Naïve Bayes (NB), and Random Forest (RF)—for motorcycle crash injury severity using data from the Highway Accident Information Management System (2020–2022). After preprocessing, 36 explanatory variables covering roadway, environmental, accident causes, crash characteristics, and vehicle involvement were analyzed. To address class imbalance, the Synthetic Minority Oversampling Technique (SMOTE) and cost-sensitive learning were applied, and models were validated using train–test splits with cross-validation. The Random Forest model achieved the best performance with an AUC of 0.726, balanced accuracy of 0.649, and Matthews Correlation Coefficient (MCC) of 0.308, outperforming the other algorithms. SHapley Additive exPlanations (SHAP) were used to interpret the RF model, identifying nighttime crashes, large truck involvement, and roadway features (e.g., depressed medians and two-lane roads) as key predictors of severe outcomes. These insights suggest countermeasures such as improving nighttime safety, dedicating truck lanes, and designing safer medians. The novelty of this study lies in integrating model comparison, imbalance-aware metrics, and SHAP interpretability to provide actionable, context-specific policy recommendations for motorcycle safety in Thailand
A Novel Exact Solution of Longshore Current and Its Application on Permeable Groin
One major environmental problem exacerbated by longshore currents is beach erosion. Groins are a common defense tactic built perpendicular to the shore. However, conventional impermeable groins promote downstream erosion and disrupt sediment movement. Permeable groins provide a more environmentally friendly option, allowing some sediment to flow through. This study examines the effects of permeable groins on longshore currents. Permeable groins are not included in currently used longshore current equations. This study fills this gap by creating a new longshore current velocity equation considering permeable groins. The longshore current equation with the groin was developed based on the momentum equation in the longshore direction without the influence of lateral mixing and the assumption that base friction will rise due to the groin. Therefore, it was determined that the base shear stress after the groin was equal to the base shear stress plus the drag caused by the groin. The result shows that the longshore current equation through the groin is a function of the breaking wave parameter and the resistance parameter owing to the groin. Longshore current velocities with and without permeable groins of different densities were measured in wave basins. We collected information on groin characteristics, current velocities, and breaking wave heights. This investigation validates the shortcomings of the current equations. Doi: 10.28991/CEJ-2025-011-02-07 Full Text: PD
Shape Functions Development for Beam-Column Element with Semi-Rigid Connections in Second-Order Steel Frame Analysis
The objective of this paper is to provide a novel method for developing the shape functions of a beam-column element with semi-rigid connection ends, thereby establishing a static analysis method for semi-rigid steel frames. This method takes into account the influence of the P-Delta effect, according to the finite element method based on displacement (FEM). The shape function is established directly from a third-order Hermitian displacement function polynomial combined with the bending element deflection differential equation. The linear elastic stiffness matrix, the geometric stiffness matrix of a semi-rigid connection beam-column, and the equilibrium equation of the element in a local coordinate system are simultaneously obtained by applying Castigliano's theorem (Part 1) for elastic deformation potential energy expression. The computational program was developed using Matlab software, and the calculation results are verified against published research results, showing that the derived shape functions and the steel frame analysis method are reliable and trustworthy. In addition, this article also derives stiffness matrices and an equivalent nodal load vector for specific cases where the semi-rigid connection is fully rigid (FR) or a pin connection. The derived shape functions are polynomial expressions with coefficients that are simply calculated from the connection stiffness and the geometric and material characteristics of the element, making them highly convenient to use. Doi: 10.28991/CEJ-2025-011-01-021 Full Text: PD
Integrating Gradient Boosting and Parametric Architecture for Optimizing Energy Use Intensity in Net-Zero Energy Buildings
Achieving net-zero energy building (NZEB) status requires accurate energy use intensity (EUI) calculations, as conventional methods often fail to capture the complexity of design and climatic conditions. In this research, a parametric energy modeling approach was used to conduct 1,350 simulations and analyze the impact of design parameters on building EUI. These simulations covered six building types”an existing building and I-, L-, T-, U-, and H-shaped buildings”across eight locations in different climate zones. A case study was conducted in Busan, Korea, where on-site measurements were obtained using portable devices to validate the simulation results. I-shaped buildings exhibited the lowest EUI, reaching 109 kWh/m²/yr at 0° and 180° orientations. The simulation results indicated that building orientations of 140°, 90°, 135°, and 270° tended to produce higher EUI values, whereas 0° and 180° showed lower EUI values of 122 and 123 kWh/m²/yr, respectively. The use of triple-pane insulated glass effectively reduced the I-shaped building's EUI to 103 kWh/m²/yr. Implementing photovoltaic (PV) systems further reduced the EUI significantly, with the I-shaped building achieving an EUI of −14 kWh/m²/yr at a 20% PV efficiency. Analysis using an extreme gradient boosting (XGBoost) model revealed that the climate zone, PV area, and type of heating, ventilation, and air-conditioning system significantly affected the EUI. This model, processed using Colab, was highly effective, with an R-squared value of 0.99, a root mean square error of 4.57, and a mean absolute error of 1.99. These findings demonstrate that the XGBoost model can effectively capture complex data patterns and can be used for high-accuracy EUI estimation. Doi: 10.28991/CEJ-2025-011-03-06 Full Text: PD
Optimizing Fire Safety and Ventilation Strategies for Structural Integrity in Rail Tunnels
Rail systems are vital to the modern urban infrastructure and offer efficient and eco-friendly transportation solutions. The Gaziray Rail System Line in Gaziantep, Türkiye addresses the region's transportation needs while considering potential hazards such as electrical malfunctions and fuel leaks. This study thoroughly assesses fire occurrences and how they affect the structural integrity of tunnel elements, thereby affecting repair costs and continuity of operations. Fire tests and modeling were employed to precisely assess tunnel fire effects, focusing on potential train fires in Gaziray Rail System Line tunnels. This study highlights the importance of vital airflow for effectively directing smoke. It also identifies the ventilation systems required to ensure optimal airflow while maintaining the structural integrity and evacuation pathways. The study identified 18 jet fans with an outlet velocity of 35.7 m/s and flow rate of 40.4 m³/s, which is essential for safe evacuation. The maximum wall temperatures ranged from 774 to 923°C, highlighting the potential fire severity. Recommendations emphasize fire-resistant materials, optimized ventilation systems, and reinforced emergency evacuation measures that are crucial for enhanced safety. Continuous training and awareness efforts ensure swift and secure evacuation during fire incidents, contributing to robust fire safety protocols for the Gaziray Rail Line. Doi: 10.28991/CEJ-2025-011-02-014 Full Text: PD
Smart Roundabout Coordination Systems for Sustainable Urban Mobility
Traffic signal coordination control is a smart approach used in urban networks to relieve the congestion by increasing corridor throughput and minimizing overall traffic delay. Previous studies have investigated various signal coordination challenges; however, integrating roundabouts into a coordinated signalized corridor without compromising their operational distinctiveness remains underexplored. This study introduces an adaptive traffic signal offset strategy incorporating a platoon compaction factor to address the dispersion effects caused by roundabouts, ensuring the preservation of platoon movement along the coordinated corridor. The method was evaluated using the PTV VISSIM micro-simulation software. The results show improvements in sustainability indicators at roundabouts, with average corridor-level delays minimized by 17%, delays associated with vehicle stops reduced by 28%, fuel consumption reduced by 16%, and emissions reduced by 9% and 16% for NOâ‚“ and CO₂, respectively. These improvements were statistically significant, affirming the robustness of the proposed method. The findings underscore the potential benefits of implementing this framework in real-world traffic scenarios, contributing to making urban transportation systems more efficient and sustainable. Doi: 10.28991/CEJ-2025-011-02-013 Full Text: PD
Mechanical Properties of Cement-Stabilized Sandy Soils Modified with Consoil
This study investigates the mechanical enhancement of sandy soils through cement stabilization modified with Consoil, targeting improved pavement substructure performance. Unconfined compressive strength (UCS) tests were conducted on samples with varying cement contents (3%, 6%, 9%), Consoil dosages (0%, 5%, 10%, 15%, 20% by cement weight), and curing periods (3, 7, 28, 90 days). Field Emission Scanning Electron Microscopy and X-Ray Diffraction analyses complemented mechanical testing to understand strengthening mechanisms. Results demonstrated that 15% Consoil consistently optimized strength development across all cement contents, with 9% cement and 15% Consoil achieving peak 90-day UCS of 17.74 MPa, representing a 67% increase over control samples. Microstructural analysis revealed progressive matrix refinement with increasing Consoil content, while XRD indicated enhanced pozzolanic activity through calcium hydroxide consumption. The study introduces Consoil as an effective stabilization additive, establishing optimal dosage rates and demonstrating significant strength improvements through synergistic cement-Consoil interactions. The findings provide new insights into strength enhancement mechanisms in Consoil-modified cement-stabilized soils, offering practical guidelines for designing high-performance pavement substructures. The research contributes to sustainable construction practices by optimizing cement usage through Consoil incorporation. Doi: 10.28991/CEJ-2025-011-01-011 Full Text: PD
An Experimental Study on Steel Fiber Effects in High-Strength Concrete Slabs
This study investigated the impact of varying steel fiber ratios by volume on the performance of HSC slabs. Incorporating steel fibers into high-strength concrete (HSC) has been shown to significantly enhance its mechanical properties, particularly by improving its load-bearing capacity. Furthermore, the addition of steel fiber reduces the reliance on traditional reinforcement bars, leading to a more efficient use of materials. This not only simplifies the construction process but also contributes to a reduction in overall construction costs. This study investigated the behavior of HSC slab specimens under loading and elevated temperatures. Three groups of specimens were created based on their thickness (8 cm, 12 cm, and 16 cm) using a single high-strength concrete mixture and four varying steel fiber proportions (0, 37.5, 75, and 150 kg/m³). Two-point monotonic loading was applied to each slab specimen until failure. To determine the splitting tensile strength, 12 cylinders were cast. Additionally, 84 cubes were cast to assess the effects of elevated temperatures and different cooling techniques on compressive strength (fcu). The results revealed that incorporating steel fibers into high-strength concrete slabs has a negligible effect on the concrete's density and compressive strength. However, it notably enhanced the splitting tensile strength and modulus of rupture. These improvements significantly boosted the material's resistance to cracking, making it more durable and better suited for applications requiring superior tensile performance. This is particularly important in structures subjected to dynamic or cyclic loading, where the risk of cracking and failure is greater. Doi: 10.28991/CEJ-2025-011-01-013 Full Text: PD
Assessment of Organic Carbon Stocks at Landscape Levels Using the InVEST Software
This study aims to calculate and assess organic carbon levels at various landscape levels of the Crimean Peninsula using the Carbon Storage and Sequestration model of the InVEST software. It outlines the stages of working with this model and highlights limitations such as the quality of input data, temporal coverage, and spatial resolution, which can significantly influence the results. Assessment of organic carbon stocks in soils, aboveground and belowground biomass, and vegetation types revealed that the highest carbon concentration was in the low-altitude landscape level of the southern macroslope. From 2017 to 2023, an annual decrease in organic carbon stocks of 0.062 t/ha was recorded, which is likely linked to climate change and shifts in land use. This research provides the first calculations of organic carbon content within the landscape levels of the Crimean Peninsula. As carbon is a significant greenhouse gas, its accumulation or emissions directly affect climate change. Evaluating organic carbon stocks in ecosystems enhances our understanding of their role in mitigating climate change and informing carbon dioxide (CO2) reduction strategies. These findings highlight the need to consider vegetation types and their changes when calculating organic carbon in landscapes and supporting regional environmental policy development. Doi: 10.28991/CEJ-2025-011-03-018 Full Text: PD
Piles Pullout Enhancement Subjected to Inclined Loads
This study focuses on the experimental and numerical analysis of pullout resistance for a single pile subjected to inclined loads in sandy soil, both before and after improvement with asphalt enhancement. The sandy soil, characterized by low cohesion, poses significant challenges for foundation stability under vertical and inclined loading conditions. Pullout tests were conducted experimentally at angles of 0°, 30°, 45°, 60°, and 90° for both vertical and horizontal components of inclined loads using a custom-designed testing setup. Cutback asphalt was introduced as an improvement agent. The experimental results revealed a significant reduction in displacement up to 62% and an improvement in pullout resistance for the asphalt-treated soil up to 55% and 72% for vertical and horizontal load directions, respectively. PLAXIS software was validated through numerical modeling, which confirmed the improved load-displacement behavior and stress distribution. The asphalt enhancement demonstrated an improvement in pullout resistance, underscoring its effectiveness in creating a cohesive soil matrix that enhances load transmission and reduces void ratios. This research provides valuable insights into the load’s inclination and improvement with angle variations; the pullout capacity enhanced significantly with the inclination angle with vertical due to the formation of a bigger failure zone, thus offering practical solutions for improving the performance of pile-supported foundations in weak sandy soils under challenging inclined load conditions