Mountain highway cut slopes subjected to excavation disturbance and rainfall infiltration often undergo progressive deformation rather than instantaneous collapse. Although previous studies have examined excavation unloading, rainfall infiltration, weak structural control, and drainage anomalies, the time–dependent interaction among these processes has not been sufficiently integrated. This review synthesizes case‐based evidence on the progressive failure of highway cut slopes under excavation–rainfall coupling. Publications from 1989 to 2026 were retrieved using Scopus and Web of Science Core Collection as the core databases, with Taylor & Francis Online used as an auxiliary publisher‐specific source and Google Scholar used for supplementary checking. After systematic screening, 95 case items representing 94 independently counted engineering cases were retained. Five non‐mutually exclusive process features were coded: staged evolution, local initiation and spatial expansion, weak‐plane or structural control, repeated sliding or reactivation, and crack‐controlled infiltration or delayed hydrological response. The results show that hydrological pathways or delayed responses, staged deformation, and structural control are recurrent features in the screened cases. The synthesized evidence indicates that excavation–rainfall coupled failure is better understood as a progressive sequence involving excavation‐induced sensitization, rainfall‐driven amplification, local initiation, deformation accumulation, and slope‐wide expansion, rather than as the simple superposition of isolated triggers. Field monitoring, laboratory testing, physical modeling, and numerical analysis provide complementary evidence for identifying stage transitions and interpreting failure mechanisms. Based on these findings, a process‐oriented framework is proposed for risk identification and early warning by linking observable slope responses with hydrological and geological conditions.
This study investigates the macro-meso mechanical evolution of dam rockfill under the coupled effects of coarse-particle content (P5) and confining pressure (σ3) using large-scale triaxial tests and PFC3D simulations. The results show (1) shear strength and particle breakage depend synergistically on σ3 and P5. For example, at P5 = 38.02%, increasing σ3 (200–800 kPa) raises Marsal’s breakage index from 3.87% to 8.64%. At a constant σ3, increasing P5 (27.56% to 70.77%) further elevates breakage from 7.41% to 9.12%. (2) Meso-kinematics quantify the granular skeleton’s transition from a “suspended” to an “interlocking” state. Higher P5 shifts the displacement field from unilateral eccentric to uniform concentric gradients, shrinking the >12 mm “long-tail” in probability density and proving that interlocking restrains extreme localized slip. (3) A critical percolation threshold emerges at P5 = 61.34%, marked by an anomalous rightward shift of the main displacement peak. This reflects global dilatancy induced by particles overriding interlocking resistance, which retracts at P5 = 70.77% as particle mobility limits lock. (4) Topological reshaping of the contact network dictates microcrack anisotropy. Higher P5 increases rigid contacts, raising total microcracks by ~6%. Consequently, crack distribution transitions from random to highly anisotropic, aligning with major force chain conjugates. Ultimately, this elucidates the failure mechanism transition from “loose slip failure” to “structural shear dislocation,” providing a scientific basis for assessing high earth-rock dam stability.
Expansive soils rich in montmorillonite minerals undergo significant volume changes with moisture fluctuations, generating swelling pressures that can adversely affect civil engineering structures. This study developed nonlinear regression models to predict the swelling pressure of expansive soils in Woldia, Ethiopia, using readily measurable soil index properties. Twenty-six disturbed and undisturbed soil samples were collected from 16 test pits (TPs) at depths of 1.5 and 3.0 m, and laboratory tests were conducted in accordance with American Society for Testing and Materials (ASTM) standards. The study’s novelty lies in the development and evaluation of one-, two-, and three-parameter nonlinear regression models using systematic data transformation and nonlinear subset-selection procedures in NCSS-2025 software. More than 214 parameter transformations were assessed to establish optimal predictive relationships between swelling pressure and soil properties. The measured swelling pressure ranged from 160 to 445 kPa and was primarily influenced by natural moisture content (NMC; 32.47%–42.85%), dry density (1.134–1.382 g/cm3), and liquidity index (−0.043 to 0.122). Among the 12 developed models, the three-parameter nonlinear model incorporating these variables provided the best predictive performance, with R2 = 0.9583, mean square error (MSE) = 0.00042, and SRMSE = 0.02058. Independent validation confirmed its reliability, yielding a prediction error of only 3.2%. The proposed model offers a rapid, reliable, and cost-effective alternative for estimating swelling pressure from routine laboratory tests, thereby reducing the need for extensive swell-consolidation testing and supporting geotechnical site investigation, foundation design, pavement engineering, and expansive-soil risk assessment in regions with similar geological and climatic conditions.
The increasing frequency of extreme rainstorms has posed significant hazards and challenges to construction management in recent years. This paper seeks to identify and categorize the current challenges and strategies encountered in construction-project management under rainstorm-disaster conditions. Relevant literature was first identified using keyword searches in two databases, Scopus and Web of Science, and then screened following the preferred reporting items for systematic reviews and meta-analyses (PRISMA) protocol. Subsequently, 83 relevant papers published between January 2000 and April 2026 were systematically analyzed and categorized into six groups, spanning pre-rainstorm construction-site disaster prediction, risk assessment and design optimization; and post-rainstorm construction-project quality inspection, schedule adjustment and safety evaluation. Additionally, the findings offer detailed insights and directions for future research from the perspectives of research objectives, data, technologies, and applications. By explicitly considering the differences between rainstorm disasters and other types of hazards, this paper provides a state-of-the-art, comprehensive reference for advanced multidimensional rainstorm-disaster management in construction projects.
Rapid urbanisation and the increasing demand for underground infrastructure have led to a substantial rise in deep excavations for basements, metro systems and utility networks. Maintaining excavation stability under complex soil and groundwater conditions remains a major geotechnical challenge because excessive ground deformation can threaten adjacent structures and public safety. This study numerically evaluates the stability and deformation behaviour of unsupported and supported vertical excavations in stratified soil using PLAXIS-2D, with emphasis on the influence of diaphragm walls, pre-stressed anchors and groundwater drawdown. A site-specific three-layer soil profile comprising fill, dense sand and clay was adopted based on laboratory-derived geotechnical parameters. Three excavation configurations were investigated under identical staged construction procedures: (i) unsupported excavation, (ii) excavation supported by a diaphragm wall and (iii) excavation supported by a diaphragm wall with pre-stressed ground anchors. The finite element model incorporated staged excavation to a depth of 9 m, groundwater lowering through progressive dewatering and asymmetric surcharge loading to simulate realistic field conditions. The responses were evaluated in terms of horizontal displacement, vertical settlement, stress redistribution, plastic strain development, pore-water pressure variation and internal forces within the support system. The unsupported excavation exhibited the highest deformation, with a maximum horizontal displacement of 87.83 mm and a maximum surface settlement of 113.85 mm. Installation of the diaphragm wall reduced the maximum horizontal displacement to 78.81 mm (10.16% reduction) and the settlement to 89.97 mm (~21% reduction). The diaphragm wall combined with two pre-stressed anchors provided the most effective support, limiting the maximum horizontal displacement to 39.78 mm (~55% reduction) and reducing the maximum settlement to 36.50 mm (~68% reduction). The anchored system also produced more uniform stress redistribution, confined plastic deformation, reduced bending demand on the diaphragm wall and maintained anchor forces within the allowable capacity. The findings demonstrate that combining diaphragm walls with pre-stressed anchors significantly enhances excavation stability by controlling deformation and improving soil–structure interaction under groundwater drawdown. The proposed numerical framework provides practical guidance for the design and assessment of supported excavations in layered soils and establishes a comparative basis for future investigations employing advanced constitutive models, coupled hydro-mechanical analyses and field-monitoring validation.
Accurate mechanical performance prediction is important for the design of lateral sliding-extension mechanism components, where slots, holes, and other geometric discontinuities can strongly affect stress concentration, deformation, stiffness, and lightweight performance. Three-dimensional finite-element analysis can provide physically meaningful mechanical labels, but constructing a sufficiently large high-fidelity dataset remains expensive because each sample requires geometry generation, meshing, numerical solution, and postprocessing. To improve prediction accuracy under limited target-domain data, this study proposes a transfer learning-based surrogate modeling framework for three-dimensional mechanical performance prediction and design assessment. A 300-sample three-dimensional finite-element dataset is generated using an open-source CalculiX workflow. Each component is described by seven design and loading variables, including geometric dimensions, slot and hole ratios, and applied force. The prediction targets include maximum displacement, maximum von Mises stress, equivalent stiffness, and a normalized design-score integrating stress, deformation, and mass-related information. Six models are compared, including Linear Regression, Random Forest, XGBoost, MLP-Scratch, MLP-FrozenTransfer, and MLP-FineTune. Two source-domain settings are further investigated: direct transfer from a general open-source mechanics dataset and transfer from a domain-aligned bridge source. The results show that source-domain alignment is critical for successful transfer learning. Under the bridge-source setting, MLP-FineTune achieves R2 values of 0.9766, 0.9967, and 0.9920 for maximum displacement, maximum von Mises stress, and design score, respectively, when 220 target-domain samples are used for training. In the small-sample case with only 20 target-domain samples, bridge-source fine-tuning reaches a design-score R2 of 0.939, clearly outperforming training from scratch. The model also preserves design ranking effectively, with a Spearman correlation of 0.9918 and a Top-5 overlap ratio of 0.88 at 220 training samples. In contrast, direct Mechanical MNIST transfer produces negative transfer in the low-data regime, indicating that general mechanics datasets may not be directly reusable for component-level design assessment. These findings demonstrate that domain-aligned transfer learning can reduce the dependence on large three-dimensional finite-element datasets and support rapid mechanical performance prediction and design screening of lateral sliding-extension mechanism components.
Dry–wet cycling is a primary cause of loess strength degradation in arid and semi-arid regions, frequently triggering geological hazards. Accurately predicting the mechanical behavior of loess under such cyclic hydraulic conditions is therefore essential for risk mitigation. While existing studies have provided phenomenological observations, the underlying micro-mechanisms remain insufficiently explained from a microphysical perspective. In this study, we propose a novel cohesive bond theory based on interparticle contact strength, distinguishing between strong bonds (from low-solubility salts) and weak bonds (from high-solubility salts). A corresponding microphysical bond-based model is developed to characterize the microscale evolution of loess cohesion. Unlike existing empirical or phenomenological models that merely fit degradation curves, our model offers a mechanistic interpretation of cohesion bond evolution, explicitly capturing both the abrupt strength drop after the first dry–wet cycle and the progressive degradation pattern with increasing cycles. To address the challenge of multivariate fitting, we introduce a high-dimensional nested fitting along with its rigorous mathematical derivation. The proposed framework is applied to predict the shear strength of Ili loess under dry–wet cycles, incorporating two key factors—dry density and number of cycles—to yield a two-dimensional function. Validation against experimental data from micropore and dry–wet cycle tests demonstrates that the model achieves high fitting accuracy, with a coefficient of determination (R2) exceeding 0.98, and effectively reproduces both the initial sharp decline and subsequent stabilization. This study provides a new theoretical framework for predicting loess behavior under hydraulic cycling, with practical value for slope stability assessment and engineering design in loess-prone regions.
To eliminate the gradation variability of reclaimed asphalt pavement (RAP) materials and improve the road performance of cold-recycled mixtures, this study analyses the gradation variability of recycled materials and conducts gradation design using the actual particle sizes of extracted aged materials. An optimisation method for the mix proportion of cold-recycled mixtures is proposed by combining the uniform method for design and the Bailey method for verification. A comparative study on the road performance of cold-recycled mixtures with two different gradations was then carried out. Based on low-temperature beam bending and semi-circular bending (SCB) tests, the low-temperature performance of the recycled mixtures was comprehensively evaluated from the perspectives of conventional analysis and energy analysis. Moreover, the correlation between fracture energy, strain energy density and the low-temperature indices specified in the current technical code was analysed to determine the low-temperature performance evaluation index for cold-recycled mixtures in severely cold regions. The results show that gradation design using only the sieving data of milled RAP materials will lead to an overall finer gradation of the recycled mixtures. For the optimised gradation verified by the uniform design and Bailey verification, the dynamic stability of the mixtures is increased by 9%~15% and the water stability is improved by 3%~6% under different cement contents. In addition, the flexural tensile strength and flexural tensile strain of the mixtures are increased by 3% and approximately 6%, respectively after gradation optimisation, indicating an improvement in low-temperature performance, with the optimal effect achieved at a cement content of 1.5%. Through the correlation analysis of fracture energy, strain energy density and flexural tensile strain, fracture energy is proposed as the low-temperature performance evaluation index for cold-recycled mixtures, and the recommended value of this index is determined to be 1000 J/m2 for severely cold regions in winter. The research results provide a theoretical basis for the gradation design and performance evaluation of recycled mixtures applied in severely cold regions.
Different Industry Foundation Classes (IFC)-to-relational mapping strategies may result in substantially different database structures and performance characteristics. However, systematic comparisons remain limited because existing studies often employ different implementation methods and evaluation conditions. To address this issue, this study developed a framework for automatically generating relational database (RDB) schemas from IFC schema definitions according to different mapping strategies and evaluating their performance under identical conditions. Three representative mapping strategies were implemented and evaluated using the BUCKY benchmark and three IFC datasets of increasing size and complexity. The results show that relational schema design has a significant impact on both database generation efficiency and query performance. Among the evaluated approaches, the inheritance-materialization strategy achieved the highest performance, while the normalized inheritance strategy provided a better balance between performance, semantic preservation, and data consistency. The proposed framework provides a reproducible methodology for evaluating IFC database architectures and building information model (BIM) server implementations.
This study evaluates the efficacy of agricultural waste—soybean pod ash (SBPA) and banana fiber (BF)—as sustainable stabilizers for expansive soils in Jimma Town, Ethiopia. Soil samples from Kebele 5 and Ajip Kela (1.50 m depth) with the highest plasticity indices and lowest California bearing ratio (CBR) values were selected. SBPA and BF were sourced locally, and various geotechnical tests assessed the natural and stabilized soils with SBPA (3%–15%) and BF (0.25%–1.25%, 15 mm length), including hybrid mixes (e.g., 3% SBPA + 0.25% BF). Sieve analysis revealed clay-rich soils: 97.34% (Ajip Kela) and 95.9% (Kebele 5) passed the number 200 sieve (0.075 mm). Classified as A-7-5 (AASHTO) and CH (Unified Soil Classification System [USCS]), the soils exhibited improved CBR and unconfined compressive strength (UCS) with SBPA-BF blends, peaking at 9% SBPA + 0.75% BF. Beyond this ratio, performance declined, indicating an optimal mix. Compaction tests showed reduced maximum dry density (MDD; 1.38–1.29 g/cm3) and increased optimum moisture content (OMC; 26.8%–31.8%), suggesting enhanced water retention and lower compaction effort. The optimum soil sample (9% SBPA + 0.75 BF) shows improved properties with reduced compression index (Cc), swelling index (Cs), and recompression index (Cr) compared to the natural soil, this indicating enhanced stiffness, lower swelling, and better stability for subgrade soil construction. The results demonstrate that SBPA-BF stabilization enhances engineering properties of soil, offering an eco-friendly, cost-effective solution for mitigating expansive soil risks. This approach aligns with sustainable waste utilization, particularly in regions like Ethiopia, where expansive soils dominate landscapes and agricultural byproducts are abundant.
To address the challenge that traditional detection methods struggle to reveal full-time-domain cumulative damage induced by temperature stress in asphalt pavements, this study analyzes the spatiotemporal evolution patterns of the asphalt pavement temperature field based on sensor monitoring data. According to these patterns, a dual-sine temperature field prediction model capable of reflecting asymmetric heating and cooling processes was established. Utilizing this model in conjunction with Miner’s linear cumulative damage theory, a complete quantitative analysis methodology translating the temperature field into a stress field and subsequently into a damage field was constructed. The results demonstrate that the dual-sine model achieves a goodness-of-fit (R2) exceeding 0.91, showcasing its unique advantage in characterizing the asymmetric daily temperature cycles in pavements. Full-time-domain damage analysis reveals that temperature-induced fatigue damage decreases significantly with depth, with annual damage degrees of 1.32 × 10−2 and 6.78 × 10−3 at depths of 0.02 m and 0.1 m, respectively. The critical source of temperature fatigue damage is identified as daily temperature differentials exceeding 18°C. Although this temperature range constitutes only about 6% of the annual duration, its damage contribution rate reaches as high as 99%. This study provides a novel, transferable method for quantifying environmental fatigue damage in asphalt pavements without requiring in-situ stress sensors, and it identifies the large-ΔT periods (spring/autumn) as the key maintenance window—a scientifically and practically significant insight for pavement design and preservation.
This study addresses the performance development and microstructural evolution of ultrahigh performance concrete (UHPC) in low-temperature environments (0–5°C) by proposing an external enhanced-insulation curing method. The effects of different curing regimes on the mechanical properties, durability, and microstructure of UHPC were systematically investigated. Two primary curing conditions representing field scenarios were designed: enhanced external insulation (C1) and noninsulated curing (C2). Standard curing (C3, 20 ± 2°C, RH > 95%) was additionally employed solely as a baseline reference for mechanical properties. Durability and microstructural comparisons were conducted exclusively between C1 and C2. The hydration temperature, compressive and axial tensile strengths, chloride-ion penetration resistance, and microstructural characteristics were evaluated by hydration-heat monitoring, mechanical tests, rapid chloride migration (RCM) measurements, thermogravimetric analysis (TGA), and mercury intrusion porosimetry (MIP). Results indicate that the enhanced-insulation condition significantly elevated the early-age hydration temperature, slowed the cooling rate, and effectively promoted cement hydration and pozzolanic reactions, thereby enhancing the mechanical performance and chloride-ion resistance. Compared with the standard-cured group, the 28-day compressive strength of the enhanced-insulation group increased by 4.5%, while the tensile strength improved by 8.1%. In contrast, the noninsulated group exhibited a marked deterioration in properties. Microstructural analyses (comparing C1 and C2) revealed that the enhanced-insulation condition promoted a higher degree of hydration, greater consumption of Ca(OH)2, and a refined pore structure, leading to a denser matrix and improved durability. The findings provide both theoretical insights and practical guidance for the engineering application of UHPC in cold regions.
Conventional polymer modification of bitumen, while enhancing in-service performance, often leads to increased material costs. Recycled tire-derived crumb rubber (CR) presents a sustainable and economically viable alternative, offering valuable elastomeric properties and environmental benefits by diverting waste from landfills. However, optimizing CR dosages and balancing its synergy with secondary polymer additives remains a critical challenge, necessitating further research to establish standardized blending ratios that maximize cost-performance efficiency without compromising viscoelastic properties under diverse climatic and traffic conditions. This study addresses this challenge by investigating the high-temperature rheological properties of polymer-modified bitumen (PMB) using the multiple stress creep recovery (MSCR) test. This robust methodology provides crucial insights into the material behavior of PMB, directly supporting the selection and development of sustainable road infrastructure. The investigation assessed the rutting resistance of four bitumen formulations at 64°C: unmodified bitumen, bitumen modified with 3% (by weight of bitumen) polyethylene (PE), 3% CR, and a combined blend of 3% PE + 3% CR. Tests were conducted at two critical stress levels (0.1 and 3.2 kPa). The findings demonstrate that the PMB sample containing both 3% CR and 3% PE exhibits superior elastic recovery and significantly reduced nonrecoverable creep compliance (Jnr) across both stress levels, indicating its enhanced resistance to permanent deformation. MSCR results showed that at 0.1 kPa, recovery increased from 2% in the unmodified binder to 10.5% with PE, 4% with CR, and 22.3% with the combined PE + CR; at 3.2 kPa, the CR + PE blend maintained 5% recovery with a Jnr of 1.63 kPa−1, outperforming PE (Jnr 5.35 kPa−1) and CR (Jnr 2.94 kPa−1). This superior performance confirms the stress-dependent behavior of PMBs, with the combined 3% CR + 3% PE blend showing the greatest improvement in rutting resistance. Statistical analysis indicates that 3% PE is suitable for standard traffic, while the CR + PE blend is preferable for heavy traffic, providing practical recommendations for performance-grade (PG) specifications.
Impact loads are common worldwide and have varying degrees of severity. Different incidents, like the sudden impact of a car accident, have significant negative consequences for infrastructure, property, and human life. Numerous studies have been conducted over the past 50 years to investigate the behavior of different structural elements with various material properties under impact load. However, limited research has been conducted on the behavior of unplasticized polyvinyl chloride (uPVC) pipes that are subjected to impact. In this study, the locally developed testing setup was calibrated against a pressure gauge up to 4136.85 kPa, producing a near-unity linear calibration coefficient of 0.99981 with an intercept of 13.494 kPa. The impact behavior of pressurized uPVC pipes under different support conditions and impactor dropping heights has been examined in this study. The pressurized pipe subjected to low-velocity impact under constant internal pressure was modelled and analyzed using ABAQUS finite element analysis (FEA) software, with a constant internal pressure of 1378.9 kPa. The plastic deformation, as well as internal pressure fluctuations due to the impact of the hammer from different heights, was investigated. For the same drop height, the peak contact force was higher under the both-end-clamped condition. The contact period of the impactor is found to be higher than the positive pulse duration of internal pressure for both support conditions.
Foundation pit engineering involves a complex system of technical standards. Yet practitioners often struggle to retrieve relevant provisions and interpret them efficiently during construction decision-making. To address these challenges, this paper presents a knowledge service system that integrates low-rank adaptation (LoRA) with retrieval-augmented generation (RAG). Based on 35 national and industry standards, we constructed a dataset of more than 8000 high-quality question–answer pairs. LoRA was adopted for parameter-efficient domain fine-tuning, and RAG was incorporated to support real-time retrieval and source attribution of regulatory provisions. In a comprehensive evaluation on 487 test items (including both subjective and objective questions), the system achieved an overall accuracy of 82.93%, gaining 16.38 and 11.15 percentage points over the pure fine-tuned model and the pure RAG model, respectively. The average response time was 1.98 s, compared with 3.65 s for the pure RAG model, thereby satisfying the real-time response requirements for on-site decision support. This study developed a prototype system using open-source frameworks, which supports offline inference and multiformat document processing in an experimental setting, thus demonstrating the feasibility and practical viability of the proposed architecture. This work offers a scalable knowledge service solution for foundation pit engineering and provides a generalizable framework for deploying large language models (LLMs) in other knowledge-intensive engineering domains.
Recycled brick aggregate (RBA) offers a practical route to reduce natural aggregate consumption, but its impact and elevated-temperature performance remain unclear when combined with hybrid fibers. This study evaluates whether rubber fiber (RF) and polypropylene fiber (PPF) can improve the impact resistance (IR) and residual compressive strength (CS) of RBA concrete. Ten mixtures were prepared using RF at 0%, 5%, and 10% replacement of fine aggregate (FA) by volume and PPF at 0%, 0.2%, and 0.4% of cement mass. The mixtures were tested for CS, splitting tensile strength (STS), flexural strength (FS), modulus of elasticity (MoE), surface absorption (SA), surface resistivity (SR), ultrasonic pulse velocity (UPV), drop-weight IR, and residual CS after exposure to 400 and 600°C. Scanning electron microscopy (SEM) and energy-dispersive spectroscopy (EDS) were used to interpret the microstructural response. The mix containing 0.4% PPF and 10% RF showed the highest IR, with a 168% increase relative to the RBA control. After exposure to 600°C, selected fiber-modified mixes showed 7%–73% higher residual CS than the heated RBA control. This residual-strength improvement was mainly attributed to matrix-level changes, pore-structure modification, and the ceramic nature of RBA, rather than intact fiber action at 600°C. SEM and EDS supported the discussion by showing changes in void size, interfacial features, and silicon-rich regions after heating. These results indicate that hybrid RF–PPF RBA concrete can improve IR while maintaining acceptable residual performance under selected elevated-temperature conditions.
This study focuses on the use of plant fibers from Hyperrhenia hirta (HH) and Stipa offneri (SO) as reinforcements in adobe production. The physical properties and durability were evaluated through an experimental campaign. The fibers were incorporated into the clay matrix at contents ranging from 0% to 1%, with a step of 0.2%. The optimal fiber contents were determined to be 0.4% for HH and 0.6% for SO. The tests show that at these contents, fiber incorporation reduces linear shrinkage from 6.87% for the reference adobe to 2.50% and 1.87% for HH-A and SO-A blocks, respectively. In contrast, it leads to a slight increase in the water absorption coefficient, rising from 0.030 kg·m−2·s−1/2 for the unstabilized adobe to 0.033 and 0.034 kg·m−2·s−1/2 for HH-A and SO-A blocks, respectively. Erosion and abrasion tests also show an improvement in durability. Mass loss due to erosion decreases from 5.72% for the reference adobe to 3.42% and 2.84% for HH-A and SO-A blocks, respectively. Similarly, mass loss due to abrasion decreases from 0.62% to 0.31% for both types of stabilized blocks. These results show that HH and SO fibers improve the dimensional stability and durability of adobe blocks. These bio-based materials thus represent a promising solution for construction in Sahelian regions. However, in areas highly exposed to water or flooding, their use should be combined with waterproof coatings to enhance durability.
Reclaimed asphalt pavement (RAP) has gained significant attention in pavement engineering due to its economic and environmental benefits. However, the high stiffness and brittleness of aged RAP binders (RPB) necessitate the use of rejuvenators to restore their performance. This study evaluated the effectiveness of two commercial rejuvenators (PX and RB) and waste engine oil (WO) at dosages of 4%, 6%, and 8% by RPB weight on the conventional, rheological, chemical, and colloidal properties of RPBs. The results showed that rejuvenator addition increased penetration from 11 dmm for the RPB to a maximum of 57.1 dmm, while reducing the softening point from 76.4 to 60.9°C and viscosity from 1610 to 1201 mPa s. The high-temperature performance grade (PG) decreased from 107.5 to 85.3°C with increasing rejuvenator dosage. Fatigue resistance improved substantially, with fatigue life increasing from 14,710 cycles for the RPB to 17,460 cycles for the PX rejuvenated asphalt binder at 2.5% strain. However, this improvement was accompanied by a reduction in rutting resistance, reflected by increased Jnr values. FTIR analysis revealed reductions in oxidation-related functional groups, with the carbonyl index decreasing from 0.0496 to 0.0343 and the sulfoxide index decreasing from 0.0935 to 0.0593, indicating partial restoration of the aged asphalt binder chemistry. SARA analysis further confirmed a reduction in asphaltene content and an increase in maltene fractions following rejuvenation. Among the investigated rejuvenators, PX exhibited better ability to restore the physical, rheological, and chemical characteristics of the aged asphalt binder under the evaluated conditions. Correlation analysis demonstrated strong relationships among conventional, rheological, chemical, and colloidal properties. Furthermore, artificial neural network (ANN), random tree (RT), and M5P models were developed to predict rutting and fatigue performance using physicochemical asphalt binder characteristics as input variables. The ANN model achieved the highest prediction accuracy, with testing correlation coefficients (CCs) of 0.9923 for Jnr and 0.9996 for Nf. Overall, the findings demonstrate that rejuvenator dosage selection requires balancing rutting resistance, fatigue performance, workability, and chemical restoration, while machine-learning techniques provide an effective framework for predicting the performance of rejuvenated RPBs.
The construction industry accounts for a substantial share of global greenhouse gas emissions, largely driven by cement manufacturing. Reducing cement consumption through strategic material substitution is therefore central to achieving more sustainable construction outcomes. One viable pathway involves identifying lightly stressed regions within structural members and replacing the conventional concrete in those zones with lower-embodied-energy alternatives, such as brick masonry or recycled aggregates, thereby cutting both material consumption and associated carbon output. This paper examines infilled RC beams as a structurally sound and environmentally responsible alternative to conventional beam design. In a standard RC beam, the tensile zone near the neutral axis carries relatively low stress; substituting this region with brick units reduces structural self-weight and lowers material cost without undermining load-carrying capacity. The extent of the substitutable zone is established through the stress-block framework of IS: 456-2000, applied in conjunction with the mechanical properties of the replacement material. Because the resulting member acts as a multilayer composite, a rigorous analytical tool is essential. The present work employs the method of initial functions (MIFs), an elasticity-based approach that avoids kinematic assumptions, to determine stress and displacement fields throughout the composite cross-section under service loading. The primary contributions of this study are threefold: (i) the sustainability-oriented concept of replacing low-stress concrete with bricks to reduce cement consumption and carbon emissions; (ii) the specific application of the MIFs to infilled RC beams as a three-layer composite system; and (iii) a quantitative assessment of stress and displacement distributions across five different replacement ratios (20%–80% and 100%). The present study extends the framework to systematically evaluate the effect of varying replacement ratios on structural performance, neutral axis migration, and composite action, providing new design-relevant data not previously available.
Structural health monitoring (SHM) of high-rise buildings is essential for intelligent infrastructure, yet conventional dynamic testing is often costly and complex to deploy, and existing roving-sensor schemes align all floor datasets to a single global time reference, leaving residual inter-session synchronisation errors that degrade mode-shape identification. To overcome these limitations, this study proposes an asynchronous vibration measurement strategy based on three-component broadband sensors. Its key innovation is a session-by-session temporal matching scheme: rather than applying one unified time reference, the roving sensor at each floor is independently synchronised to a fixed top-floor reference through an improved frequency-domain decomposition (FDD) method, eliminating the synchronisation errors introduced by independent data acquisition. Using only three sensors, the method non-invasively characterises a 26-storey reinforced-concrete frame–shear wall building; because each three-component sensor simultaneously records the east–west (E–W), north–south (N–S) and vertical (Z) responses, natural frequencies and mode shapes in all three directions are obtained from a single campaign. The identified first- and second-order natural frequencies are 0.75 and 3.20 Hz horizontally and 3.66 and 7.50 Hz vertically; the results are highly consistent across all sessions and agree with empirical estimates to within about 10%. The proposed strategy provides a reliable and low-cost three-directional modal baseline for the building’s initial healthy state, supporting lifecycle SHM, damage warning and intelligent operation and maintenance.
