Search

The objective of this study is to create explicit prediction models for the complex shear modulus (G*) and phase angle (δ) of bitumen mastic fabricated using an evolutionary machine learning approach. The dynamic shear rheometer (DSR) test in frequency sweep mode at seven test temperatures was performed to measure G* and δ. In order to create specific prediction models for each modifier, multigene genetic programming (MGGP) was employed. These models took into account various factors including the dosage of the additive, filler volume filling rate, loading frequency, temperature, as well as the G* and δ values of the neat bitumen. In general, six explicit prediction models are presented for different additives with R-squared values of more than 0.9. The results showed that the hybrid machine learning approach can effectively develop precise, meaningful, and yet simple formulas for calculating G* and δ of the bitumen mastic. To gain a deeper understanding of the developed models, a comprehensive parametric study and sensitivity analysis were carried out. [ABSTRACT FROM AUTHOR]

• Predicting J nr and %R of modified bitumens deterministically by using MGGP based on MSCR test results. • Predicting probability of different traffic levels by using logistic regression technique. • Determining the critical dosages of crumb rubber, SBS, and PPA modifiers that alter the traffic levels. This study aims to develop deterministic and probabilistic prediction models for the multiple stress creep and recovery (MSCR) test. For this purpose, crumb rubber, polyphosphoric acid, and styrene–butadiene–styrene bitumen modifiers have been used with different dosages to modify high-temperature performance of PG 58–28 and PG 64–22 base bitumens. The MSCR test has been performed at different temperatures. Deterministic models are developed by the multi-gene genetic programming technique for each modifier individually, and the non-recoverable creep compliance (J nr) and percent recovery (R) parameters are predicted. The accuracy of deterministic models is suitable and the performance of R models has been better than J nr models. Furthermore, a comprehensive probabilistic model has been developed by using the logistic regression technique to predict different traffic levels. The accuracy of the probabilistic model is 0.85. The sensitivity analysis has been performed on this model and the effect of changes in the modifier dosage and temperature on the traffic levels have been investigated. Results show that using the probabilistic model, it is possible to find a range of modifier's dosage in which the traffic level is desired. [ABSTRACT FROM AUTHOR]