Recommendations to Improve Quality of Safety Indicators in the Railway Industry
Bibliographic record
Abstract
Summary & ConclusionsDespite the overwhelming number of currently available safety management systems, accidents still periodically occur in the railway industry. This paper analyses recent incidents and accidents in the Canadian railway industry, showing that in many of these cases, identifiable precursors were present in the trajectory of these accidents. Several repeating precursors can be identified from the railway accident reports in the Transportation Safety Board of Canada (TSB) database. Examples include damage to non-critical safety equipment, operator complaints, maintenance problems, quality problems, or, more importantly, similar past incidents for which either the underlying causes were not correctly identified, or the mitigation measures were not satisfactorily implemented.These recurring precursors and patterns of precursors can be seen as warning signals. Their existence indicates a gap between the actual status of a system with regard to safety and the common proactive safety indicators. This gap consists of information, already present and available in the industry in some form, but unavailable to the system safety analyst; it results in situations where safety indicators, such as risk analyses, do not accurately represent the system under consideration. This paper argues that these reoccurring precursors, were they included in the safety indicators, would provide a clearer picture of the actual safety deficiencies of the systems and aid in the prevention of similar accidents.This paper will primarily focus on analyzing the data from "unplanned/uncontrolled movement of rolling stock" occurrences from main track or sidings, in order to find opportunities for further enhancing safety reporting, management, and performance. Uncontrolled movements are relatively rare events, which, despite their low probability of occurrence, can have catastrophic consequences—particularly if the rolling stock involved are carrying dangerous goods and are unattended. The Lac-Mégantic rail accident of 2013 demonstrated that the cost to human life and our communities can be incalculable.Despite significant safety action taken by Transport Canada and the railway industry since the Lac-Mégantic accident to reduce the probability of unplanned/uncontrolled movements of rail equipment, this type of occurrence has continued to trend upwards, posing a significant risk to the rail transportation system. The increase in these occurrences is particularly important in light of the fact that the amount of dangerous goods transported by rail within Canada has increased by an average of approximately 25% since 2004, with a 42.5% increase in transported fuels and chemicals between 2011 and 2017. Further, movement of dangerous goods by rail is forecasted to continue increasing. Sustainable growth in the transport of dangerous goods by rail will require acceptable safety levels.It is argued in this paper that one of the main contributors to the problem of repeated similar incidents and accidents originates from the fragmentation of the railway industry. The rail industry now operates as a complex web of different operating companies, infrastructure management companies, regulatory bodies, and contractors. Even though this fragmentation is understandable and even beneficial in supporting the wide range of freight and light rail applications internationally, inconsistent reporting of accidents and incidents impedes the improvement of safety, industry wide.We believe that a more centralized and integrated incident reporting system similar to those in industries like Nuclear or Aviation could improve the understanding of potential risks, the design of new systems, and guide regulation. The data from such reports, available in a publicly accessible database, could be used for validating safety analyses, improving quantitative analyses, and to lower precursor frequency through informed design. As a result, the probability of more serious accidents may be reduced, and the safety of daily operation of railway systems improved.
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 | 0.081 | 0.207 |
| Meta-epidemiology (narrow) | 0.003 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.003 |
| Bibliometrics | 0.010 | 0.010 |
| Science and technology studies | 0.003 | 0.002 |
| Scholarly communication | 0.014 | 0.012 |
| Open science | 0.008 | 0.005 |
| Research integrity | 0.008 | 0.008 |
| Insufficient payload (model declined to judge) | 0.024 | 0.010 |
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".