Akademik Veri Yönetim Sistemi
2nd International Conference on Information Technologies and Their Applications, ITTA 2024, Baku, Azerbaycan, 23 - 25 Nisan 2024, cilt.2226 CCIS, ss.3-17
In this study, the optimization of rutting and fatigue performance characteristics in SBS and geopolymer composite-modified bitumen was achieved by using an experimental design approach and an Artificial Neural Network (ANN) model. The Design Expert 10.0.3 software was employed to construct an optimal mixture design, taking into account the ratios of SBS and Geopolymer within the bitumen matrix. A reference binder of B 50/70 penetration grade bitumen was used, and modified binders were prepared considering various ratios of SBS and Geopolymer. The factors are defined as SBS and Geopolymer concentrations, and the responses include SHRP rutting (G*/sin δ) and fatigue (G*sin δ) factors, non-recoverable creep compliance (Jnr), and number of cycles to failure (Nf), which were measured through Dynamic Shear Rheometer (DSR), Multiple Stress Creep Recovery (MSCR), and Linear Amplitude Sweep (LAS) tests. In experimental design, four central composite design (CCD) models were constructed, and variance analysis was performed to investigate the statistical significance of the model results. In ANN model, the multilayer perceptron was employed. Findings indicated that a higher percentage of Geopolymer in SBS modified bitumen is considered a promising option for reducing the SBS content, thus offering a potential pathway for the enhancement of the performance characteristics of the modified bitumen.