File(s) not publicly available
Complexity on the rails: A systems-based approach to understanding safety management in rail transport
journal contribution
posted on 2019-08-22, 00:00 authored by GJM Read, Anjum NaweedAnjum Naweed, PM Salmon© 2019 Rail transport is a complex, safety-critical system which experiences catastrophic events. Systems theory-based approaches have been applied to understand risk and safety management in a range of safety-critical domains, however in rail, systems approaches have tended to be applied for accident analysis, rather than to describe how risk and safety is proactively managed. This study involved the development of a control structure model of rail transport in Australia using the Systems Theoretic Accident Model and Processes (STAMP) method. The control structure identifies the actors involved in managing safety, and shows the control and feedback mechanisms that comprise the adaptive feedback function to maintain safety. The control structure was refined in a participative process involving subject matter experts. Insights identified for improving safety management included: a need to improve feedback mechanisms to better understand the effectiveness of control measures; a lack of formal controls at higher levels of the system; and a focus within current feedback mechanisms on failures rather than understanding and learning from normal performance. Opportunities for applying the model in industry practice are also explored.
Funding
Category 1 - Australian Competitive Grants (this includes ARC, NHMRC)
History
Volume
188Start Page
352End Page
365Number of Pages
14eISSN
1879-0836ISSN
0951-8320Publisher
Elsevier, UKPublisher DOI
Peer Reviewed
- Yes
Open Access
- No
Acceptance Date
2019-03-19External Author Affiliations
University of the Sunshine CoastAuthor Research Institute
- Appleton Institute
Era Eligible
- Yes
Journal
Reliability Engineering and System SafetyUsage metrics
Categories
Keywords
Licence
Exports
RefWorksRefWorks
BibTeXBibTeX
Ref. managerRef. manager
EndnoteEndnote
DataCiteDataCite
NLMNLM
DCDC