Evaluating Natural Language Processing Algorithms for Improved Hazard and Operability Analysis
Bibliographic record
Abstract
Automating Hazard and Operability (HAZOP) reports is crucial for enhancing efficiency and reducing human biases in hazard identification process. Recent HAZOP studies have applied various Natural Language Processing (NLP) algorithms to optimize hazard identification and reporting. These studies exhibit divergent textual inputs and results and overlook the mining of all HAZOP report components. Furthermore, traditional NLP algorithms such as Bag of Words (BOW) and Term Frequency-inverse Document Frequency (TF-IDF) often fail to capture polysemy and semantic relationships between words. This study aims to evaluate NLP algorithms’ efficacy in automating HAZOP reports to improve safety levels in infrastructure projects. These algorithms, including BOW, TF-IDF, the global vectors for word representation (GloVe), and sentence bidirectional encoder representation from transformers (SBERT), were combined with machine learning classifiers such as random forest (RF), gaussian naive bayes (NB), decision tree (DT), and k-nearest neighbors (KNN), using two small HAZOP datasets with varied inputs. A zero-shot text classification model was further evaluated for its ability to assign labels to HAZOP data without prior training. Results demonstrate that GloVe combined with RF achieves the highest accuracy (83 %), significantly outperforming other models. We further observe that KNN degrades on short text features, while DT underperforms on longer descriptions. The zero-shot model achieves low performance (52 % accuracy), lacking the precision needed for fine-grained, jargon-heavy labels. These findings indicate that GloVe embeddings remain a robust foundation for HAZOP automation under data-scarce conditions in the operations of infrastructure projects.
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 | 0.006 | 0.016 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.003 | 0.002 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.003 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.001 |
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".