Association Rules Mined from Construction Accident Data
- Title
- Association Rules Mined from Construction Accident Data
- Author
- 서종원
- Keywords
- fatality; serious injury; association rule; data mining; construction accident
- Issue Date
- 2018-05
- Publisher
- KOREAN SOCIETY OF CIVIL ENGINEERS-KSCE
- Citation
- KSCE JOURNAL OF CIVIL ENGINEERING, v. 22, no. 4, page. 1027-1039
- Abstract
- Worker safety awareness issue in construction job site is a major concern due to the hazardous work conditions attributed to the dynamic and complex nature of construction job sites. Providing intuitive knowledge expressed as linguistic statements would be useful for corrective and/or preventive actions. This study investigates 98,189 serious injury and fatal accidents that occurred in Korean building construction sites in the period 2006-2010 in order to discover intuitive knowledge expressed as association rules among multi-attributes of construction accidents. 74 association rules are identified as valid by computing the minimum support, confidence, and lift of each rule. Finally, 30 consolidated association rules are obtained as meaningful knowledge for implementing corrective and preventive actions for effective job site safety management after either combining or pruning those 74 rules. This study provides the theoretical rule base that defines the association among accident attributes causing serious injuries and fatalities at construction job site. These rules may be useful for efficient safety control and education.
- URI
- https://link.springer.com/article/10.1007%2Fs12205-017-0537-6https://repository.hanyang.ac.kr/handle/20.500.11754/118776
- ISSN
- 1226-7988; 1976-3808
- DOI
- 10.1007/s12205-017-0537-6
- Appears in Collections:
- COLLEGE OF ENGINEERING[S](공과대학) > CIVIL AND ENVIRONMENTAL ENGINEERING(건설환경공학과) > Articles
- Files in This Item:
There are no files associated with this item.
- Export
- RIS (EndNote)
- XLS (Excel)
- XML