Prediction of asphalt rheological properties for paving and maintenance assistance using explainable machine learning
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
• Laboratory results show correlations between | G* | and δ and chemical properties. • Influential factors in predicting rheological properties are investigated. • The superior performance of the FT-XGBoost model is validated. • Explanations for machine learning predictions are included. Conventional frequency-temperature sweep tests for evaluating asphalt rheological properties are time-consuming and resource-intensive. The characterization efficiency can be significantly improved by establishing a robust predictive model that links rheological properties to chemical composition. To this end, this study investigates the correlation between asphalt’s chemical and rheological properties and develops precise predictive models using machine learning techniques. The input features include eleven key functional groups measured by Fourier Transform Infrared Spectroscopy (FTIR), while the output variables are the complex modulus (| G* |) and phase angle ( δ ) from Dynamic Shear Rheometer (DSR). Five machine learning algorithms—multiple linear regression, support vector regression, artificial neural network, random forest, and eXtreme gradient boosting (XGBoost)—were utilized to construct the predictive models. A Bayesian optimization strategy was employed to fine-tune their hyperparameters. Laboratory findings revealed that a strong correlation was identified between changes in these functional groups, especially oxygen-containing functional groups, and the | G* | and δ values of asphalt binders. The optimized XGBoost model achieved exceptional predictive accuracy, with R 2 values of 0.9998 for | G* | and 0.9999 for δ . Additionally, SHapley Additive exPlanations (SHAP) values were used to elucidate the underlying principles of the predictions. By leveraging FTIR data and rheological indicators, this work provides a novel data-driven approach to accurately estimate asphalt binder behaviour, reducing experimental effort while ensuring reliable performance evaluation.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it