Asian Journal of Civil Engineering最新文献

{"title":"Performance evaluation of fiber-optimized self-compacting concrete: mechanical, microstructural, and life cycle perspectives","authors":"Ramamohana Reddy Bellum,&nbsp;Talapa Reddy Suman Kumar,&nbsp;Chava Venkatesh","doi":"10.1007/s42107-026-01678-8","DOIUrl":"10.1007/s42107-026-01678-8","url":null,"abstract":"<div><p>The construction industry requires sustainable high-performance materials, as cement production contributes approximately 8% of global CO₂ emissions. This study presents a fiber-optimized high-strength self-compacting concrete (SCC) incorporating fly ash and silica fume as partial cement replacements, focusing on rheological behavior, mechanical performance, microstructural evolution, and life-cycle sustainability. A total of 40% cement replacement was achieved using 160 kg/m³ fly ash and 40 kg/m³ silica fume, while polypropylene fibers were incorporated at 0.02%, 0.04%, 0.06%, and 0.08% by volume to identify the optimum fiber dosage that balances flowability and strength. Fresh properties were evaluated using slump flow, L-box, and V-funnel tests, while mechanical properties were assessed through compressive, split tensile, and flexural strength tests at 7, 14, and 28 days. Microstructural characteristics were examined using XRD, FTIR, TGA/DTA, and SEM–EDS analyses. The results demonstrate that an optimal fiber dosage of 0.06% yields superior performance, achieving compressive, split tensile, and flexural strengths of 74.85 MPa, 6.46 MPa, and 5.79 MPa, respectively, representing improvements of 8.3%, 11.0%, and 6.8% over the control mix. Microstructural analysis confirms enhanced C–S–H formation, pore refinement, and effective fiber bridging. Life-cycle assessment quantified environmental performance across six impact categories, revealing that the integration of 40% supplementary cementitious materials reduced global warming potential by 34.6% (from 791.83 to 517.86 kg CO₂-eq/m³) and primary energy demand by 31.3% (from 8643.38 to 5938 MJ/m³) relative to conventional concrete. Fiber-reinforced variants exhibited minimal environmental variation across all categories—global warming potential ranged 517.07–517.86 kg CO₂-eq/m³, acidification potential 4.580–4.589 kg SO₂-eq, and eutrophication potential 0.744–0.754 kg PO₄³⁻-eq. Economic evaluation demonstrates that the optimal mix achieved production costs of ₹7366.08/m³ with superior cost-performance ratio of 105.53 ₹/MPa compared to 123.60 ₹/MPa for conventional concrete, representing 14.6% improved efficiency. The study establishes an optimized SCC formulation that combines high mechanical performance, quantified environmental benefits, and economic viability for sustainable construction applications.</p></div>","PeriodicalId":8513,"journal":{"name":"Asian Journal of Civil Engineering","volume":"27 7","pages":"3477 - 3497"},"PeriodicalIF":0.0,"publicationDate":"2026-05-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148386103","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Improving the reliability of nondestructive concrete strength assessment in new structures using machine learning techniques","authors":"Khoudja Ali-Benyahia,&nbsp;Mohamed Ghrici,&nbsp;Said Kenai,&nbsp;Ilyas Ali-Benyahia","doi":"10.1007/s42107-026-01701-y","DOIUrl":"10.1007/s42107-026-01701-y","url":null,"abstract":"<div><p>Accurate estimation of concrete strength in new structures is critical for quality control. Coring methods provide reliable results but are impractical for large-scale applications. In practice, it is impossible to test the strength of all the concrete to be cast in a structure. Testing the strength of selected samples, as commonly practiced in many countries, is insufficient for making reliable decisions about in-situ concrete quality. Non-destructive testing (NDT) methods offer effective alternatives. In this study, the strengths of selected samples are correlated with NDT measurements on corresponding structural elements, from which the potential strength is indirectly estimated at any location within the structure. The rebound hammer (RH) and ultrasonic pulse velocity (UPV) are widely used but require well-calibrated models. Conventional calibration approaches, typically based on linear regression, often fail to capture the complex, nonlinear, and multivariable nature of real-world data. This study explores the use of machine learning (ML), specifically Random Forest Regression, to enhance predictive accuracy. A dataset of 448 in-situ triplets (cube compressive strength, RH, and UPV) collected from three building sites during construction is analyzed. Complementary parameters such as concrete age, curing conditions, and element type, often neglected in traditional models, are incorporated to better capture nonlinear relationships. An extensive evaluation using statistical performance indicators, including root mean square error, prediction risk, prediction error variability, and the coefficient of determination, combined with Monte Carlo simulations, enables a probabilistic performance assessment, demonstrating that the ML-based model outperforms traditional regression methods in both accuracy and reliability. These findings highlight the potential of artificial intelligence to enhance the reliability of NDT-based diagnostics and prognostics for concrete strength estimation.</p></div>","PeriodicalId":8513,"journal":{"name":"Asian Journal of Civil Engineering","volume":"27 8","pages":"3809 - 3828"},"PeriodicalIF":0.0,"publicationDate":"2026-05-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148458936","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Bayesian-optimized explainable machine learning framework for predicting reinforced concrete beam ultimate load-carrying capacity","authors":"Arvind Dewangan,&nbsp;Nikita Jain,&nbsp;Neha Sharma,&nbsp;Sagar Paruthi,&nbsp;Rupesh Kumar Tipu","doi":"10.1007/s42107-026-01692-w","DOIUrl":"10.1007/s42107-026-01692-w","url":null,"abstract":"<div><p>Accurate prediction of ultimate load-carrying capacity (failure load) of reinforced concrete (RC) beam is important for structural assessment, preliminary design, retrofit planning and decision support. Conventional empirical and code-based approaches are still useful, but they have a tendency to reduce the complexity of the nonlinear interaction of geometry, reinforcement, and material properties. This study proposed a Bayesian-optimized machine learning framework for predicting RC beam ultimate load-carrying capacity (failure load) using tabular engineering data and also converting the final predictor in the form of a readable graphical user interface. The dataset included 3234 records of the beam with nine input variables representing the beam geometry, concrete grade, ratio of longitudinal reinforcement, stirrup details, steel yield strength and concrete compressive strength with the target response being the ultimate load-carrying capacity (failure load) in kN. Seven regression models, that is, Linear Regression, Support Vector Regression, Decision Tree Regressor, Random Forest Regressor, Gradient Boosting Regressor, Extreme Gradient Boosting and Multilayer Perceptron were trained and compared using 80:20 train-test split and 5-fold cross validation with Hyperparameter optimization (Bayesian Optimization) Model performance was assessed in terms of <span>(R^2,)</span> RMSE, MAE, and MAPE and SHAP analysis was employed for the interpretation of the best-performing model. The results indicated that the Multilayer Perceptron had the highest predictive accuracy with the test <span>(R^2)</span> of 0.9992, RMSE of 21.3059 kN, MAE of 17.0777 kN, and MAPE of 0.4570%, which was better than all the competing models. Gradient Boosting Regressor and Extreme Gradient Boosting appeared to be the second best group while the Decision Tree Regressor showed the weakest generalization. SHAP interpretation showed that the most important predictors were longitudinal reinforcement ratio, beam width, beam height, concrete compressive strength and concrete grade with approximately 99.73% of global importance in the final model. The developed graphical user interface serves as a proof-of-concept tool to demonstrate the practical usability of the trained machine learning model for rapid prediction of reinforced concrete beam ultimate load-carrying capacity. The interface is intended primarily for preliminary assessment, educational demonstration, and exploratory analysis, rather than as a direct substitute for code-based structural design procedures. The results of the study show that a carefully optimized and interpretable machine learning framework can provide highly accurate and practically deployable prediction of RC beam ultimate load-carrying capacity (failure load) for structural engineering applications.</p></div>","PeriodicalId":8513,"journal":{"name":"Asian Journal of Civil Engineering","volume":"27 8","pages":"3747 - 3773"},"PeriodicalIF":0.0,"publicationDate":"2026-05-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148458935","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Design optimization and performance prediction of biocomposite based brick using machine learning techniques","authors":"Vidula Ganesh Waskar,&nbsp;Shrishail B. Anadinni","doi":"10.1007/s42107-026-01694-8","DOIUrl":"10.1007/s42107-026-01694-8","url":null,"abstract":"<div>\u0000 \u0000 <p>The construction sector’s 40% contribution to global CO₂ emissions underscores urgent demand for bio-based masonry alternatives to energy-intensive cement/clay units. This study pioneers mycelium-bound sugarcane bagasse bio-composite (BC) Brick, leveraging agro-waste (70–85 vol%) and Pleurotus ostreatus via systematic bio-fabrication—optimizing PSD (C/M/F: 55/30/15→40/40/20), spawn (5–10 wt%), growth (7–14 days), and uniaxial pressing (0.15–0.4 MPa). Comprehensive assessment yields: bulk density 268–490 kg/m³ (+ 7.5–9.6% maturation), CS 0.18–1.85 MPa (40–50%), water absorption 74–162% (-12-14%), sorptivity 0.012–0.038 mm/√min (-14-19%), thermal conductivity 0.026–0.048 W/m·K (-7-10%)—M12 optimum (70% bagasse, 40/40/20 PSD, 10% spawn, 0.4 MPa) via hyphal capillary occlusion and closed-cell refinement. Response surface methodology excels (R²&gt;0.997, RMSE &lt; 0.33; Adj.R²&gt;0.998) for quadratic optimization, outperforming ANN (3-10-4; R²=0.327, RMSE = 0.289) amid biological nonlinearities. Cradle-to-gate LCA evidence 60–80% GWP abatement (32 kg CO₂ eq/m³ vs. 180 clay bricks), driven by ambient processing/waste upcycling. This novel experimental-ML-LCA paradigm positions mycelium BC as a scalable, non-structural insulation, catalysing the transition of fungal materials toward decarbonised construction.</p>\u0000 </div>","PeriodicalId":8513,"journal":{"name":"Asian Journal of Civil Engineering","volume":"27 8","pages":"3715 - 3732"},"PeriodicalIF":0.0,"publicationDate":"2026-04-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148458808","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Axial compressive behavior of square CFST short columns with steel plate reinforcement","authors":"Ayce Cathen Winti,&nbsp;Abdul Azim Abdullah,&nbsp;Abdul Razak Abdul Karim,&nbsp;Ahmad Beng Hong Kueh,&nbsp;Bethany Guenevere Bronny","doi":"10.1007/s42107-026-01658-y","DOIUrl":"10.1007/s42107-026-01658-y","url":null,"abstract":"<div><p>This study examines the axial compressive behavior of square concrete filled steel tube (CFST) short columns enhanced with steel plate reinforcement (SPR). Although CFST columns benefit from composite action, square sections exhibit reduced confinement effectiveness for the concrete core, increasing susceptibility to local buckling. While various strengthening methods are available, most of their implementation with prefabricated steel tubes is frequently impractical due to fabrication complexities that adversely affect construction efficiency. To overcome these issues, nine 100 mm <span>(times )</span> 100 mm <span>(times )</span> 300 mm square CFST short column specimens were tested experimentally under axial compression, including unreinforced, single SPR and double SPR configurations. SPR elements were spot welded internally to the steel tube walls to improve local stability with minimal fabrication complexity. Results confirm that SPR effectively delays local buckling onset, redistributes stress concentrations and enhances concrete confinement. Compared to unreinforced columns, specimens with SPR demonstrated increases in initial elastic stiffness (up to 63%), ductility (up to 39%) and ultimate axial compressive strength (up to 25%). Notably, double SPR configurations provided higher strength and energy absorption capacity, whereas single SPR delivered superior initial elastic stiffness and ductility, indicating a nonlinear performance relationship with reinforcement quantity. For analytical convenience, design equation is proposed to predict the axial compressive load capacity of square CFST short columns with SPR. These outcomes highlight SPR’s viability as a simple, efficient strengthening alternative, offering enhanced structural performance and constructability over established methods.</p></div>","PeriodicalId":8513,"journal":{"name":"Asian Journal of Civil Engineering","volume":"27 6","pages":"3099 - 3122"},"PeriodicalIF":0.0,"publicationDate":"2026-04-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s42107-026-01658-y.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147959104","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Unified fuzzy–opposition-based NSGA-III framework for multi-objective time–cost–quality–safety optimization in construction scheduling","authors":"Om Prakash Singh,&nbsp;Niraj Kumar Singh,&nbsp;Subhash Chandra Bhushan,&nbsp;Madan Kumar,&nbsp;Sandeep Kumar Sahoo,&nbsp;Mani Bhushan,&nbsp;Jagannath Padhy","doi":"10.1007/s42107-026-01681-z","DOIUrl":"10.1007/s42107-026-01681-z","url":null,"abstract":"<div>\u0000 \u0000 <p>Construction projects are inherently affected by uncertainties arising from site variability, resource availability, and subjective scheduling judgments. This study proposes a unified hybrid framework that integrates Fuzzy Logic, Opposition-Based Learning (OBL), and the Non-Dominated Sorting Genetic Algorithm III (NSGA-III) to address multi-objective trade-offs involving Time, Cost, Quality, and Safety (TCQS) under uncertainty. Triangular Fuzzy Numbers (TFNs) are used to model uncertain parameters such as activity durations, direct costs, quality indices, and safety risks. These are defuzzified using the centroid method to enable effective optimization. A real-world construction case study comprising 21 activities, each with five execution modes, is used to demonstrate the approach. The proposed Fuzzy-OBL-NSGA-III algorithm yields 25 high-quality Pareto-optimal solutions, balancing the conflicting TCQS objectives. Performance metrics and comparative analysis with existing models such as MOACO, MOTLBO, MODE, standard NSGA-III, and Fuzzy-MOPSO confirm the superior diversity, convergence, and computational efficiency of the proposed model. Trade-off plots and correlation analyses further validate its decision-support capabilities. The framework provides construction planners with a dynamic, scalable, and uncertainty-resilient scheduling tool, enabling strategic decision-making in complex, real-life project environments.</p>\u0000 </div>","PeriodicalId":8513,"journal":{"name":"Asian Journal of Civil Engineering","volume":"27 7","pages":"3529 - 3546"},"PeriodicalIF":0.0,"publicationDate":"2026-04-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148386040","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Predictive modeling of compressive strength in silica fume–modified RCC using ML under organic acid attack","authors":"Abhishek Mishra,&nbsp;Mukesh Pandey,&nbsp;Rakesh Gupta","doi":"10.1007/s42107-026-01680-0","DOIUrl":"10.1007/s42107-026-01680-0","url":null,"abstract":"<div>\u0000 \u0000 <p>The degradation of reinforced cement concrete (RCC) due to organic acids poses critical challenges for infrastructure longevity, particularly in agro-industrial, dairy, and wastewater environments. This study investigates the compressive strength performance of silica fume–modified RCC exposed to citric acid and proposes a machine learning (ML) framework for predictive modeling. A total of six mixes incorporating 0% to 15% silica fume were experimentally evaluated for compressive strength at 7, 28, and 56 days. Post-curing, samples were exposed to citric acid concentrations ranging from 0% to 10% to simulate organic degradation. Experimental results showed that 10% silica fume significantly enhanced strength and acid resistance. A synthetic dataset (<i>n</i> = 1000) was generated using controlled domain-informed sampling across seven input variables. Five ML models—Support Vector Regression, Decision Tree, Extra Trees, Random Forest, and XGBoost—were trained and validated using R², RMSE, MAE, MAPE, a20, and IOA. XGBoost achieved the highest performance (R² = 0.9782; MAPE = 2.81%). SHAP and PDP interpretability techniques identified curing age, silica fume content, and acid concentration as critical features. The proposed model demonstrates strong predictive accuracy and robustness, offering a reliable tool for optimizing RCC design in acid-exposed environments.</p>\u0000 </div>","PeriodicalId":8513,"journal":{"name":"Asian Journal of Civil Engineering","volume":"27 7","pages":"3511 - 3527"},"PeriodicalIF":0.0,"publicationDate":"2026-04-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148386043","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An explainable XGBoost model for predicting cost overruns in construction projects","authors":"Minh-Thu Nguyen,&nbsp;Van-Tien Phan,&nbsp;Ha-Lan Tran,&nbsp;Minh-Anh Nguyen Vu","doi":"10.1007/s42107-026-01686-8","DOIUrl":"10.1007/s42107-026-01686-8","url":null,"abstract":"<div><p>Cost overrun is one of the most prevalent challenges in construction projects, significantly affecting investment efficiency and project management performance. As construction projects become increasingly large-scale and complex, the development of accurate predictive models plays a crucial role in supporting cost risk management. This study proposes a machine learning based approach for predicting construction project cost overruns using the Extreme Gradient Boosting (XGBoost) algorithm integrated with the SHAP (SHapley Additive exPlanations) interpretability technique. The dataset consists of 1000 simulated observations, incorporating key characteristics of construction projects, including project size, estimated project cost, material costs, schedule pressure, delay risk, design changes, and macroeconomic indicators such as the construction material price index and inflation rate. The model was optimized using Randomized Search combined with 5-Fold cross-validation to enhance predictive performance. Model accuracy was evaluated using Mean Absolute Error (MAE) and the coefficient of determination (R²). The results demonstrate that the XGBoost model achieves high predictive accuracy and effectively identifies the most influential factors contributing to cost overruns. Furthermore, SHAP-based analysis enhances model interpretability, providing valuable insights to support decision-making in construction cost management.</p></div>","PeriodicalId":8513,"journal":{"name":"Asian Journal of Civil Engineering","volume":"27 7","pages":"3611 - 3628"},"PeriodicalIF":0.0,"publicationDate":"2026-04-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148386044","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Machine learning-based prediction and optimization of mechanical and durability properties of geopolymer concrete","authors":"U. R. Kawade,&nbsp;Prital Kalasur,&nbsp;B. S. Karthik,&nbsp;Praveen Oggu,&nbsp;Ravikumar Seenivasan,&nbsp;N. Lingeshwaran","doi":"10.1007/s42107-026-01669-9","DOIUrl":"10.1007/s42107-026-01669-9","url":null,"abstract":"<div>\u0000 \u0000 <p>In this study, an integrated experimental and machine learning–based framework was developed to predict and optimize the fresh, mechanical, and durability properties of basalt fiber–reinforced ground GGBS based GPC. A total of 100 systematically designed mix proportions were prepared by varying NaOH (12–17%), Na₂SiO₃ (28–37%), activator-to-binder ratio (0.40–0.53), liquid-to-binder ratio (0.30–0.38), and BF dosage (0–1.0%). Experimental results indicated that the 28-day CS ranged from 52.3 to 85.3 MPa, TS from 5.5 to 9.0 MPa, and FS from 6.8 to 10.9 MPa. Durability performance was characterized by water absorption of 3.25–4.85%, sorptivity of 5.05–7.85 × 10⁻⁷ mm/√s, and rapid chloride penetration resistance of 980–1850 Coulombs at 28 days, while UPV values exceeding 4.5 km/s confirmed excellent internal matrix quality. RSM models demonstrated strong predictive capability with R² values between 0.87 and 0.99, enabling multi-response optimization. An ANN model developed using a 5–10–8 architecture achieved an overall RMSE of 0.204, MAE of 0.157, and R² of 0.242 under aggregated multi-output evaluation, highlighting its ability to capture complex nonlinear behavior. Optimal performance was consistently obtained at activator-to-binder ratios of 0.46–0.50, liquid-to-binder ratios of 0.32–0.36, and BF contents of 0.50–0.75%. The proposed experimental ML framework provides a robust decision-support tool for designing high-performance and durable geopolymer concrete, supporting its application in sustainable construction.</p>\u0000 </div>","PeriodicalId":8513,"journal":{"name":"Asian Journal of Civil Engineering","volume":"27 6","pages":"3299 - 3322"},"PeriodicalIF":0.0,"publicationDate":"2026-04-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147959105","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A metaheuristic–machine learning framework for modeling and improving the thermal behavior of bio-based wall panel systems in residential buildings","authors":"Tariq Qarem,&nbsp;Rehab Salaheldin Ghoneim,&nbsp;Mazin Arabasy","doi":"10.1007/s42107-026-01668-w","DOIUrl":"10.1007/s42107-026-01668-w","url":null,"abstract":"<div>\u0000 \u0000 <p>Bio-based wall panel solutions are being promoted as sustainable alternatives to conventional wall construction; however, the thermal performance of these solutions is influenced by complex interplays between material properties, geometric configurations, and environmental conditions. This paper proposes an integrated machine learning and metaheuristic approach to predict and optimize the thermal, economic, and environmental performance of bio-based wall panel solutions for residential buildings. The heterogeneous dataset was developed by incorporating experimental data, validated simulations, and literature data to train various predictive models. The accuracy of the advanced nonlinear machine learning models was found to be high (R² up to 0.85), which was better than the accuracy of the linear machine learning model and proved the existence of interaction-driven thermal behavior. The results of the explainability analysis identified wall thickness as the primary factor influencing the performance of the wall panel solutions, followed by porosity and moisture-related attributes. The optimization results quantitatively determined the trade-offs between thermal transmittance, embodied carbon, and cost and proved that low U-value wall panel solutions can be developed with moderate embodied carbon and cost levels. The results of this research demonstrate that it is possible to attain high levels of thermal efficiency with relatively low embodied carbon levels, especially for well-designed plant fiber wall panel solutions.</p>\u0000 </div>","PeriodicalId":8513,"journal":{"name":"Asian Journal of Civil Engineering","volume":"27 6","pages":"3285 - 3297"},"PeriodicalIF":0.0,"publicationDate":"2026-04-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147959167","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}