Chemical process safety domain knowledge graph‐enhanced <scp>LLM</scp> for efficient emergency response decision support
Bibliographic record
Abstract
Abstract Chemical process safety accidents are characterized by their sudden onset, rapid evolution, and severe consequences. Developing effective emergency response decisions for such complex and dynamic incidents requires comprehensively considering various knowledge domains. Relying solely on expert experience and emergency plans often fails to meet the demands of effective emergency response. To enhance the efficiency of emergency response decision‐making in chemical process accidents, this study proposes a method that leverages a chemical process safety knowledge graph (CPSKG) to enhance large language models (LLMs) for generating reliable emergency response decisions. The proposed method uses a seven‐step approach to designing scenario and emergency response ontologies. By aligning with the characteristics of emergency domain knowledge texts and the ontology framework, natural language processing (NLP) and retrieval‐augmented generation using graphs (Graph RAG) techniques are employed to construct a semantically rich CPSKG. The entities and relationships within the graph enhance the reasoning capabilities of LLMs, facilitating the generation of efficient and reliable emergency response decisions. A case study was conducted to validate the reliability of this approach. The results demonstrate that the LLM enhanced with the CPSKG outperforms other models in generating more effective emergency response decisions. As a key contribution, the proposed method improves the efficiency of knowledge sharing and emergency response in the chemical process safety domain while generating reliable and auxiliary emergency decisions.
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 | 0.001 | 0.004 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.002 | 0.002 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.005 | 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".