Forecasting Marshall stability of waste plastic reinforced concrete using SVM, ANN, and tree-based techniques

• Published in: Multiscale and Multidisciplinary Modeling, Experiments and Design Vol. 7; no. 4; pp. 4569 - 4587
• Main Authors: Kumar, Bhupender, Kumar, Navsal
• Format: Journal Article
• Language: English
• Published: Cham Springer International Publishing 01.09.2024
• Subjects: Characterization and Evaluation of Materials; Engineering; Mathematical Applications in the Physical Sciences; Mechanical Engineering; Numerical and Computational Physics; Original Paper; Simulation; Solid Mechanics; Marshall stability; Random forest; Artificial neural network; Support vector machine; Random tree
• ISSN: 2520-8160; 2520-8179
• DOI: 10.1007/s41939-024-00501-8

Improving the durability and performance of asphalt concrete is crucial in pavement engineering, given its fundamental role as a foundational material. This study utilizes advanced soft computing methods to predict the Marshall stability (MS) of asphalt concrete enhanced with waste plastic, aiming to enhance its effectiveness and longevity. By compiling a comprehensive dataset from various experimental investigations and reputable scholarly sources, the study evaluates a range of models, including artificial neural network (ANN), random forest (RF), random tree (RT), support vector machine (SVM), and bagging RT. These models are meticulously assessed across seven statistical metrics to gauge their ability in MS prediction, including coefficient of correlation (CC), root mean square error (RMSE), mean absolute error (MAE), relative absolute error (RAE), root relative squared error (RRSE), mean absolute percentage error (MAPE), scatter index (SI), and comprehensive measure (COM). Among these models, the random forest (RF) model emerges as the most effective, displaying a notable CC value of 0.8719. Its minimal RMSE, MAE, RAE, and RRSE values further indicate its precision in forecasting MS. Particularly noteworthy is the sensitivity analysis conducted on the RF model, revealing the significant impact of aggregate size on MS prediction and highlighting the model's ability to identify crucial factors influencing asphalt concrete stability. This research holds substantial implications for pavement engineering, providing valuable insights into enhancing the quality of asphalt concrete through the incorporation of waste plastic reinforcement. By leveraging advanced computational methodologies, the study offers a reliable framework for MS prediction, thereby advancing the development of more resilient and environmentally sustainable pavement materials. The findings underscore the potential of waste plastic integration while emphasizing the importance of considering aggregate size in optimizing asphalt concrete stability. Overall, this research contributes significantly to enhancing infrastructure sustainability and effectiveness on a broader scale.