Assessment of the Long-Term Impacts of Highway–Railway Grade Crossing Countermeasures: A Bayesian Vector Autoregression Modeling Approach
Bibliographic record
Abstract
This paper proposes integrated machine learning and time-series models to investigate the long-term impact of a variety of safety countermeasures on the consequence score of train-vehicle collisions at highway–railway grade crossings (HRGCs). To that end, train-vehicle collisions that occurred at HRGCs in the United States between 2009 and 2018 were extracted from the US Federal Railroad Administration (FRA) collision dataset, along with the countermeasures that are implemented at each HRGC location. The consequence score of each collision was extracted from a web-based platform named GradeDec.Net, administrated by the FRA. A nonlinear M5Prime (M5P) model tree was developed to classify the collision dataset into a set of homogeneous classes based on the characteristics of the HRGCs (namely, train speed, highway speed, and traffic volume). In total, the HRGCs considered in the study were classified into five classes based on the results of the M5P model. Then, a Bayesian vector autoregression (BVAR) model was developed for each class to understand the temporal trends of the safety impact of eight countermeasures on the consequence score of collisions. The study showed that the impact of several countermeasures fluctuates over time depending on the prevailing conditions of the HRGCs. Some countermeasures showed negative short-term impacts in some classes, but in the long run, their safety benefits become evident. Other countermeasures showed limited short-term benefits but in the long run, their safety benefits deteriorate significantly. Moreover, the forecasting accuracy of the proposed BVAR model was evaluated by comparing the model predictions to the observed consequence score in the three years following the period used to train the model (2019–2021).
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 machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.004 | 0.006 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.003 | 0.002 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.001 | 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 source (direct Gemma or distilled Codex), 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".