Hybrid machine learning-enabled multivariate bridge-specific seismic vulnerability and resilience assessment of UHPC bridges
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
Efficient seismic vulnerability and resilience assessment is essential for ultra-high-performance concrete (UHPC) bridges, given their distinctive mechanical and structural properties. However, existing single-parameter-based probabilistic seismic demand (PSD) models overlook critical bridge‐specific characteristics and uncertainties. Besides, studies on seismic vulnerability and resilience assessment of UHPC bridges are scarce. Thus, this study proposes a hybrid machine learning (ML)-enabled multivariate bridge-specific seismic vulnerability and resilience assessment framework for UHPC bridges. Key design parameters and associated uncertainties are identified, and a Latin Hypercube Sampling (LHS) technique is employed to establish a representative UHPC bridge database, which is used to develop a hybrid ML model-based multivariate PSD model. A comparative analysis with the conventional PSD model, as well as widely used ML algorithms, demonstrated that the proposed PSD model achieves the highest predictive performance, characterized by the highest coefficient of determination and lowest prediction errors. Additionally, SHapley Additive exPlanation (SHAP) analysis is used to investigate the effect of different parameters on the PSD of UHPC bridges. The results of SHAP show the peak ground acceleration (PGA) as the most important factor, followed by bridge span and column diameter. The hybrid ML-enabled multi-variate bridge-specific fragility analysis results are used to investigate the functionality recovery and resilience of the bridge, which demonstrate the reduction in the residual functionality and overall bridge resilience with the increase in the ground motion intensity.
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