Abstract

Flexible pavements are subject to progressive deterioration driven by repeated traffic loading, environmental exposure, and material degradation, which causes distress. This study presents an integrated research framework that combines laboratory performance evaluation of rice straw fiber (RSF)-modified hot mix asphalt with machine learning-based pavement distress prediction to advance sustainable pavements. Three plant-produced HMA mixtures were modified with untreated and alkali-treated RSF at dosage levels of 0.2%, 0.4%, and 0.6% by total mixture weight. FTIR spectroscopic analysis confirmed that alkali treatment induced significant chemical modifications to the fiber surface, increasing hydroxyl group availability and enhancing fiber–binder adhesion. Alkali-treated RSF consistently outperformed untreated RSF in rutting resistance across all mixture types, with the 0.2% treated dosage achieving rut depth reductions of up to 47% relative to the unmodified control. Untreated RSF generally provided higher CT-Index and fatigue index (Sapp) responses, whereas alkali-treated RSF improved rutting resistance and generally reduced DR; therefore, the preferred RSF treatment is distress- and mixture-dependent. Prediction models for rutting, fatigue cracking, transverse cracking, and IRI were developed using the LTPP SPS-10 database and seven machine learning (ML) algorithms. XGBoost provided the best rutting prediction (overall R² = 0.96), LightGBM provided the best fatigue cracking (overall R² = 0.92) and transverse cracking (overall R² = 0.88) prediction, and Artificial Neural Network provided the best IRI (overall R² = 0.96) prediction. Implementation of the developed models demonstrated that different RSF dosages are projected to produce different long-term distress trajectories under different traffic conditions. The integration of laboratory fiber characterization with ML-based 7 distress prediction establishes a unified, data-driven framework for evaluating the feasibility of agricultural waste valorization in sustainable pavement engineering.

Date of publication

Summer 7-13-2026

Document Type

Thesis

Language

english

Persistent identifier

http://hdl.handle.net/10950/5136

Share

COinS