Machine Learning–Based Decision Support Framework for Construction Injury Severity Prediction and Risk Mitigation
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
Construction is a key pillar in the global economy, but it is also an industry that has one of the highest fatality rates. The goal of the current study is to employ machine learning in order to develop a framework based on which better-informed and interpretable injury-risk mitigation decisions can be made for construction sites. Central to the framework, generalizable glass-box and black-box models are developed and validated to predict injury severity levels based on the interdependent effects of identified key injury factors. To demonstrate the framework utility, a data set pertaining to construction site injury cases is utilized. By employing the developed models, safety managers can evaluate different construction site safety risk levels, and the potential high-risk zones can be flagged for devising targeted (i.e., site-specific) proactive risk mitigation strategies. Managers can also use the framework to explore complex relationships between interdependent factors and corresponding cause-and-effect of injury severity, which can further enhance their understanding of the underlying mechanisms that shape construction safety risks. Overall, the current study offers transparent, interpretable and generalizable decision-making insights for safety managers and workplace risk practitioners to better identify, understand, predict, and control the factors influencing construction site injuries and ultimately improve the safety level of their working environments by mitigating the risks of associated project disruptions.
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 | 0.004 | 0.002 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.002 |
| 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