Integrating Extreme Vertices Design and Genetic Algorithm to Enhance the Moisture Resistance of Bituminous Concrete Modified with Recycled High-Density Polyethylene

This study combined Extreme Vertices Design (EVD) and Genetic Algorithm (GA) to improve the moisture resistance of HDPE-modified bituminous concrete by enhancing its tensile strength ratio (TSR). EVD was employed for mixture design, sensitivity analysis, and model development using regression techniques, which were validated through F-statistics, R², and mean absolute percentage deviation (MAP.D). GA was used to optimize the mixture components to further enhance the TSR. The TSR of HDPE-modified mixtures ranged from 82.73% to 95.82%, surpassing the ASTM and AASHTO thresholds, while the unmodified mixture had a TSR of 82.43%. HDPE-modified mixtures met or exceeded the 80% TSR requirement, with several exceeding 85%. The analysis identified key interactions, such as the positive effect of granite content on TSR and the negative impact of excess sand and bitumen. Optimal HDPE content was found to improve TSR up to a certain point, beyond which it reduced the mixture's cohesion. ANOVA indicated significant interactions between granite, bitumen, and HDPE in influencing TSR. The regression model demonstrated high predictive capability with an R² of 93.27% and a low MAP.D of 0.74%, confirming its adequacy. GA optimization yielded a TSR of 97.88%, significantly improving upon the unmodified mixture's performance. The optimized mixture, with specific proportions of granite, sand, bitumen, and HDPE, surpassed ASTM and AASHTO standards, showing the effectiveness of HDPE in enhancing both performance and sustainability in asphalt mixtures.