Development of petroleum-derived polymeric additive to enhance the bituminous properties with the use of a machine-learning model

Abstract

Study revealed the synthesis of petroleum-derived polymeric additive PS-co-PAO. The synthesis of the additives was confirmed by FT-IR and NMR spectroscopy. The synthesized polymeric additives was blended with VG 10 base bitumen to modifiy its physicochemical and rheological properties. Six modified bitumens SOMB1,SOMB2, SOMB3, SOMB4, SOMB5, and SOMB6 were prepared using different concentrations of polymeric additives at 150°C and 170°C. These modified bitumens were further analyzed for the physicochemical and rheological properties which revealed that the SOMB-6 modified bitumen was found most suitable. Additionally, a Random Forest regression model was developed to predict the rutting resistance based on the percentage of polymeric additive and temperature. The model demonstrated a high R-squared value of 0.986, indicating that it can effectively predict rutting resistance, enhancing the design and testing of polymer-modified bitumen. A study on multifunctional additives also revealed that the prepared modified bitumens marginally meet the properties of higher-grade VG and modified bitumen as per IS and IRC specifications.

Keywords

• Bitumen
• Polymer-modified bitumen
• Rheology
• Rutting resistance
• Viscosity grade bitumen
• Machine learning