Dataset for Predicting Asphalt Fatigue Life: Investigating Material and Loading Parameters with a Comprehensive Dataset and Addressing Model Intricacies within Weave-UNISONO 2021 project, NCN project No 2021/03/Y/ST8/00079, and GACR project GA22-04047K

Summary: This study employs artificial neural networks (ANNs) to predict the fatigue life of asphalt concrete (AC). Leveraging a dataset from extensive laboratory tests, the ANN model was optimized to address the variability in AC fatigue data. Our approach involved fine-tuning hyperparameters and adapting network architectures to best utilize a dataset of 152 samples (two laboratories CUT Prague & WUT Warsaw). Additional parameters and tuning code for the ANN model is available on GitHub https://github.com/jakub-houlik/Asphalt_Fatigue_ANN The dataset includes: Data collection includes initial stiffness (MPa) and number of cycles (times) to fatigue (4PB) at loading frequency 10 Hz and test temperature 10°C. Outcomes of the experimental carried out on ACP 22+ 50/70, ACL 16+ 50/70, ACO 11+ 50/70, ACP 16+ 60RA, ACP 22S 50RA and 70/100, VMT 22 30RA and PMB 25/55-65RC, ACL 16+ 50RA and 70/100, ACL 16+ 50RA and 70/100, ACL 16S 50RA and PMB RC, ACL 22S 50RA and PMB RC, ACL 22S 50RA and PMB RC, ACO 11+ 30RA, AC16W PMB25/55-60, AC16W 35/50, AC16W 50/70 mixtures: 01 asphalt_10_deg.csv