SAFETYLIT WEEKLY UPDATE

We compile citations and summaries of about 400 new articles every week.
RSS Feed

HELP: Tutorials | FAQ
CONTACT US: Contact info

Search Results

Journal Article

Citation

Long D, Wei J, Yang C. China Saf. Sci. J. 2021; 31(12): 10-16.

Copyright

(Copyright © 2021, China Occupational Safety and Health Association, Publisher Gai Xue bao)

DOI

10.16265/j.cnki.issn1003-3033.2021.12.002

PMID

unavailable

Abstract

In order to address insufficiency of traditional construction behavior risk analysis in reflecting problems in projects and regions due to subjective judgement of experts, a knowledge graph method was proposed in this article for improving construction behavior-based safety risk analysis and hazardous location identification. Firstly, based on objective data report of construction accidents, an accident knowledge graph was built to store and recall data efficiently as well as support reasoning. Then, accident data deriving from construction behavior was extracted from the graph to modify possibility degree function of risk grading initially defined by experts, and the method based on gray clustering for risk analysis of construction behavior was improved. Finally, path reasoning algorithm over the knowledge graph was formed to identify relevant hazardous location corresponding to construction behavior indicators. After key behavior indicators were determined based on risk analysis results. Furthermore, a knowledge graph was developed through a case study on the same type of recent construction accident reports in the region where the project was located, the behavior-based safety risk analysis and hazardous location identification were carried out. The results show that the proposed method can determine accident level dynamically and comprehensively, and reflect risk location and timing characteristics, thus helping improve construction safety measures. © 2020 China Safety Science Journal. All rights reserved.


Language: zh

NEW SEARCH


All SafetyLit records are available for automatic download to Zotero & Mendeley
Print