The Safety Science Entity Atlas is an open, interactive inventory of the models, theories, methods and tools used in safety science and patient safety. Each of its 181 coding units is drawn as a schematic diagram and coded on two axes: the field it belongs to (seven fields, A to G) and the kind of entity it is (twelve types).
How it was built
Corpus: a title-level Web of Science search, screened for relevance, giving 1,809 records (1,657 articles and 152 reviews, 1980 to 2026). A supplementary search with tool, instrument and standard terms added 671 screened records and five units. Units: 146 single entities and 35 lineages that merge versions and sector variants of one core. Of the 181 units, 133 are observed in the main corpus, 8 only in the supplementary search, and 40 are canonical, known from the reference literature only. Coding: field, conceptual family, primary type, usage role, level of analysis, temporal orientation and primary function, refined over six documented critique rounds with the decision questions shown in the Types view. Field records where an entity originated, not where it is applied. Corpus prominence: the number of screened records that name a unit in the title, abstract or keywords. It shows how visible an entity is in publications that name models, not how often it is used, and it is a lower bound for measurement instruments, protocols, decision tools and standards. How far to trust the coding
Two independent coders who took no part in developing the scheme recoded a stratified sample of 55 units, and disagreements were settled by the author. Field, primary type, level of analysis and primary function were reproduced well. Temporal orientation was reproduced only moderately and usage role poorly, even after the rules were clarified, so read those two as descriptive notes. Full statistics are reported in the accompanying manuscript.
What goes beyond the taxonomy study
The inventory, the coding dimensions, the type definitions, the list of precursors and the six propositions come from the taxonomy study. The schematic diagrams, lineage member lists, influence network, guided tours, model selection guide and Incident Room are editorial and teaching layers that the author added for this atlas. They are selective, have not been coded independently, and should not be cited as findings.
Use of AI
Large language model assistants were used to implement the extraction pipeline, to apply the written screening rule to records with generic search terms (checked by the author on a blind sample), to draft and critique successive versions of the coding scheme, and to draw the diagrams and build this interface. All coding decisions were reviewed and approved by the author, who takes full responsibility for the content. Corrections and additions are welcome through the project repository .
How to cite
Kurutkan, N. (2026). Safety Science Entity Atlas (Güvenlik Bilimi Varlık Atlası), inventory v10 [Data set]. Zenodo. https://doi.org/10.5281/zenodo.23138415
Copy citation
Author and licence
Prof. Dr. Mehmet Nurullah Kurutkan, Düzce University, Faculty of Business, Department of Health Management. ORCID 0000-0002-3740-4231 . Licensed under CC BY 4.0 : share and adapt with attribution.
The diagrams are schematic drawings for teaching and do not replace the original sources. Incident files in the Incident Room are fictional.