Soft computing-based optimization of plastic waste utilization in flexible pavement construction

• Published in: Multiscale and Multidisciplinary Modeling, Experiments and Design Vol. 7; no. 4; pp. 3087 - 3098
• Main Authors: Kumar, Bhupender, Kumar, Navsal, Kashyap, Veena
• 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; Review; Simulation; Solid Mechanics; Flexible pavement; Predictive modeling; Sustainable infrastructure; Machine learning; Plastic waste
• ISSN: 2520-8160; 2520-8179
• DOI: 10.1007/s41939-024-00399-2

The increasing interest in employing plastic waste in the construction of flexible pavements has been driven by growing environmental concerns and the demand for sustainable infrastructure solutions. This research explores the application of machine learning techniques to enhance the efficient use of plastic waste in the design and prediction of flexible pavement performance. Utilizing a dataset derived from previous literature, predictive models were constructed using a variety of machine learning algorithms, encompassing artificial neural networks, random trees, and support vector machines. The performance of these models was rigorously evaluated using three key statistical metrics: correlation coefficient (CC), mean absolute error (MAE), and root-mean-square error (RMSE). The evaluation results show that the artificial neural networks-based model outperformed the other models in precisely predicting the Marshall stability of asphalt concrete containing plastic waste, with CC values of 0.8422 and 0.829, MAE values of 1.8335 and 1.6516, and RMSE values of 2.3582 and 2.1336 for both training and testing stages. Furthermore, a sensitivity analysis conducted within this study emphasized the pivotal role of plastic size as a critical parameter, particularly when plastic waste is introduced into the mix. This underscores the significance of optimizing plastic size in such scenarios.