Identifying the emerging vulnerability of railway transport systems across countries by automated analysis of railway accident reports
Bibliographic record
Abstract
Although railway accident reports and recommendations are proposed after railway accidents, practitioners and researchers suffer from the need to deal with a large amount of textual data given that most railway safety-related information is recorded and stored in the form of text. Hence, there is a growing need for accurate estimations of the vulnerability of railway transport and for effective mitigation strategies. This thesis extends knowledge on the vulnerability of the railway system by exploring the underlying hazards and building rigorous and automated models to enlarge the database. The conceptual frameworks HazardMap and RecoMap were developed to overcome this gap, using Natural Language Processing (NLP) topic models for the automated analysis of textual data to extract critical insights. Empirical data was retrieved from official railway accident reports published by four countries: Australia - the Australian Transport Safety Bureau (ATSB), the UK - Rail Accident Investigation Branch (RAIB), the US - National Transportation Safety Board (NTSB) and Canada - the Transportation Safety Board of Canada (TSB). Scoping workshops and a survey were conducted to evaluate the usefulness and consistency of railway practice. Case studies of the application to the risk at level crossings and the platform–train interface risks are provided to illustrate how the models proposed work with real-world data. The interpretation of findings indicates the potentially emerging hazard of deterioration in railway safety. Potential barriers to learning across jurisdictions and time might deteriorate the organisational safety culture and endanger railway. To address such obstacles, the HazardMap and RecoMap proposed are capable of automating hazard analysis with adequate accuracy to help stakeholders better understand hazards and help practitioners learn across jurisdictions and time.
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 | 0.008 | 0.038 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.015 | 0.012 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.003 | 0.003 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| 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".