Prediction of Asphalt Performance Based on Plastic Waste Using Machine Learning - Open Access Article

Aug 5, 2026

Incorporating plastic waste into asphalt mixtures provides a practical pathway to reduce environmental burdens while improving road material performance. Traditional approaches such as the Marshall test are expensive and time-consuming, creating a strong demand for faster prediction methods. Machine learning models including random forest, XGBoost, and artificial neural networks can predict asphalt performance, optimize mixture composition, and lessen dependence on labor-intensive laboratory testing by leveraging key factors like bitumen content, plastic size, and temperature. This review supports sustainable, cost-effective pavement development and future innovations.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages10

Reading Time25 min

AccessFree to download

Article Info

Abstract

1. INTRODUCTION

Prediction of asphalt performance based on plastic waste using machine learning I Gusti Agung Ananda Putra1, Darpan Rokade2 1Environmental Engineering, Faculty of Engineering and Informatics, Universitas Pendidikan Nasional, Denpasar, Indonesia 2Computer Science, Faculty of Computer Science, P.E. S. Modern College of Arts, Science and Commerce, Pune, India | Article history:
Received Apr 29, 2025 Revised Aug 19, 2025 Accepted Sep 27, 2025 | The incorporation of plastic waste into asphalt mixtures offers a promising solution to address the growing environmental concerns while enhancing the performance of road materials. Traditional methods, such as the Marshall test, are costly and time-consuming, thus highlighting the need for more efficient prediction techniques. Machine learning (ML) models, including random forest (RF), extreme gradient boosting (XGBoost), and artificial neural networks (ANN), have shown significant potential in predicting asphalt performance, optimizing material compositions, and reducing the dependence on labor-intensive laboratory tests. Key influencing factors such as bitumen content, plastic size, and temperature have been identified as crucial for improving asphalt properties. This systematic review emphasizes the potential of ML in streamlining the development of plastic-modified asphalt, offering a sustainable and cost-effective approach to road construction. Furthermore, it supports the advancement of green infrastructure and lays the foundation for future innovations in sustainable pavement engineering, contributing both to academic research and practical applications in the construction industry.
This is an open access article under the CC BY-SA license.
| | --- | --- | | Keywords:
Asphalt mixtures Machine learning Performance prediction Plastic waste
Sustainable road construction | | Corresponding Author: I Gusti Agung Ananda Putra Environmental Engineering, Faculty of Engineering and Informatics, Universitas Pendidikan Nasional Denpasar, Indonesia Email: [anandaputra@undiknas.ac.id](anandaputra@undiknas.ac.id) Article Info ABSTRACT 1. INTRODUCTION The global issue of plastic waste has reached a concerning level, with annual production exceeding 359 million tonnes, causing serious environmental and health issues [1] . The magnitude of plastic pollution is further highlighted by studies showing that a significant portion of this waste enters oceans, exacerbating ecological damage [2] . The road construction sector emerges as a potential solution for utilising plastic waste by incorporating it as a reinforcement material in asphalt, thereby reducing environmental burdens while enhancing material performance [3] . Early demonstrations in countries like India have validated the technical feasibility of plastic roads, showcasing enhanced durability under tropical conditions [4] . However, the variability in the physicochemical properties of plastics and the uncertainties surrounding their interactions with bitumen present challenges in designing optimal mixtures [5] . Chemical incompatibilities between certain plastics and bitumen can lead to phase separation, necessitating rigorous pre-treatment protocols [6] . Empirical studies demonstrate that incorporating waste plastics such as polyethylene terephthalate (PET), low-density polyethylene (LDPE), and high-density polyethylene (HDPE) into hot mix asphalt can enhance Marshall stability and flow properties, indicating improved mechanical performance with increasing plastic content [7] . Field trials confirm that polyethylene-modified asphalt exhibits superior resistance to cracking compared to conventional mixes [8] . Furthermore, the conventional Marshall method for estimating bitumen ratios is costly and time-consuming [9] . Recent advances propose machine learning (ML) as a viable alternative to empirical methods, reducing testing time [10] . Consequently, there is a pressing need for more efficient and accu

FAQs

Why is predicting asphalt performance important for plastic-modified mixtures?

Plastic-modified asphalt aims to improve material properties while addressing plastic waste concerns, but variability in plastic properties and interactions with bitumen makes design difficult. Accurate prediction helps achieve reliable mixture performance without excessive trial-and-error.

Which machine learning models are highlighted for predicting asphalt performance?

The document highlights random forest (RF), extreme gradient boosting (XGBoost), and artificial neural networks (ANN) as effective approaches for predicting asphalt performance and optimizing mixture compositions.

What factors most strongly influence asphalt properties in plastic-modified research?

Key influencing factors include bitumen content, plastic size, and temperature, which are identified as crucial for improving asphalt properties and performance predictions.

Aug 5, 2026

Upload by Theodora