Raising the bar in safety performance is a critical challenge for many organizations. Contributing factor taxonomies organize information on why accidents occur. Therefore, they are essential elements of accident investigations and safety reporting systems. Organizations must balance efforts to identify causes of specific accidents with efforts to evaluate systemic safety issues in order to become more proactive about improving safety. This research successfully addressed two problems: (1) limited methods and metrics exist to support the design of effective taxonomies, and (2) influence relationships between contributing factors are not explicitly modeled within a taxonomy. The primary result of the taxonomic relationship modeling efforts was an innovative "dual role" contributing factor taxonomy with significant improvements in comprehensiveness and diagnosticity over existing taxonomies. The influence map methodology was the result of a unique graphical and analytical combination of the dual role taxonomy and influence relationship models. Influence maps were developed for several safety incidents at Kennedy Space Center. An independent assessment was conducted by a team of experts using the new dual role taxonomy and influence chain methodology to evaluate the accuracy and completeness of contributing factors identified during the formal incident investigations. One hundred and sixteen contributing factors were identified using the influence map methodology. Only 16% of these contributing factors were accurately identified with traditional tools, and over half of the 116 contributing factors were completely unaddressed by the findings and recommendations of the formal incident reports. The new methodology is being applied to improve spaceport operations and enhance designs of future NASA launch systems.
Identifer | oai:union.ndltd.org:ucf.edu/oai:stars.library.ucf.edu:etd-1960 |
Date | 01 January 2006 |
Creators | Barth, Timothy |
Publisher | STARS |
Source Sets | University of Central Florida |
Language | English |
Detected Language | English |
Type | text |
Format | application/pdf |
Source | Electronic Theses and Dissertations |
Page generated in 0.0181 seconds