Soft computing applications in asphalt pavement: A comprehensive review of data-driven techniques using response surface methodology and machine learning

Abstract

The asphalt pavement industry is transforming because of the growing influence of artificial intelligence and industrial digitization. As a result of this shift, there is a stronger emphasis on advanced statistical approaches like optimization tools like response surface methodology (RSM) and machine learning (ML) techniques. The goal of this paper is to provide a scientometric and systematic review of the application of RSM and ML applications in data-driven approaches such as optimizing, modeling, and predicting asphalt pavement performance to achieve sustainable asphalt pavements in support of numerous sustainable development goals (SDGs). These include Goals 9 (sustainable infrastructure), 11 (urban resilience), 12 (sustainable construction strategies), 13 (climate action through optimized materials), and 17 (multidisciplinary interaction). A thorough search of the ScienceDirect, Web of Science, and Scopus databases from 2010 to 2023 yielded 1249 relevant records, with 125 studies closely examined. Over the last thirteen years, there has been significant research growth in RSM and ML applications, particularly in ML-based pavement optimization. The study shows that the topic has a global presence, with notable contributions from Asia, North America, Europe, and other continents. Researchers have concentrated on utilizing sophisticated ML models such as support vector machines (SVM), artificial neural networks (ANN), and Bayesian networks for prediction. Also, the integration of RSM and ML provides a faster and more efficient method for analyzing large datasets to optimize asphalt pavement performance variables. Key contributors include the United States, China, and Malaysia, with global efforts focused on sustainable materials and approaches to reduce impact on the environment. Furthermore, the review demonstrates the integrated use of RSM and ML as transformative tools for improving sustainability, which contributes significantly to SDGs 9, 11, 12, 13, and 17. Providing valuable insights for future research and guiding decision-making for soft computing applications for asphalt pavement projects.