Hybrid generative adversarial network and machine learning approach for performance prediction of marshall stability and marshall flow of recycled asphalt shingle pavements
Bibliographic record
Abstract
Recycling asphalt shingles waste offers a promising way to enhance the performance of hot-mix asphalt pavements and reduce both environmental impact and reconstruction costs. Growing interest in sustainable pavement materials has initiated research into how recycled asphalt shingle (RAS) can improve key performance indicators such as Marshall Stability (MS) and Marshall Flow (MF). Despite this impetus, progress has been delayed due to limited and imbalanced datasets, making it difficult for traditional machine learning models to deliver reliable predictions. As a result, much of the potential of RAS remained unrealized, and decision-makers lacked robust tools to evaluate performance outcomes. Therefore, this study introduced a conditional tabular generative adversarial network to generate synthetic data from 70% of the real dataset. The method produced a well-balanced training environment for model development, which helped overcome data paucity and improve prediction accuracy. This hybrid strategy significantly improved the generalization of the model and accuracy by achieving R² values of 0.9887 for MS and 0.9505 for MF for the testing set, respectively. The Huber losses of MS and MF of the testing sets were 1.654 and 0.0763, in the stated order. SHAP analysis further revealed that binder content and mix gradation were the most influential parameters, reinforcing existing domain knowledge while opening the door to more data-driven pavement design. This work not only filled a critical data gap but also demonstrated the powerful integration between synthetic data and advanced ML models in civil engineering applications.
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.
How this classification was reachedexpand
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 itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".