Date of Award
Doctor of Philosophy (PhD)
Shappell, Scott A
Gramopadhye , Anand
Melloy , Brian
Stringfellow , Paris
Wiegmann , Doug
Many models of accident investigation have been created and have served as the basis for other tools and techniques. One of the most prominent techniques, Reason's Swiss Cheese Model (1990), is based on the idea of active and latent failures. Expanding on this idea, the Human Factors Analysis and Classification System (HFACS) was created with the idea to associate contributing factors and errors (Wiegmann & Shappell, 2003).
While HFACS has been frequently applied in the field of aviation in literature, other industry types are under-represented. Seventeen data sources encompassing various industry types were collected and included in this dissertation analysis. While each industry type is unique, the human constituent is a shared element among industries. A multi-industry analysis will allow for common high-level human error patterns to emerge and for benchmarking standards to be created. It is also important to identify relationships between active errors and latent conditions without limiting data to one specific industry type while concurrently using a taxonomy that systemically identifies both active errors and latent conditions at all levels of an organization. Doing so could potentially allow for the shifting of intervention target areas from active errors to latent conditions and for assistance in identifying other potential errors and latent failures during investigation.
As a result, four sets of benchmarking standards were established, and a decision support tool was created to assist in selecting the correct benchmarking standard set. Additionally, twelve adjacent tier causal factor associations and three additional non-adjacent tier causal factor associations were found to be significant. Due to the tiers' ease of investigation and classification, most associations were between the preconditions for unsafe act tier and preconditions for unsafe act tier.
Overall, this dissertation furthers the research field of HFACS and its application. Originating in Reason's (1990) Swiss Cheese Model, the HFACS taxonomy aims to identify the holes in the Swiss cheese. This dissertation furthered progress in determining the size of the holes and the interactions among the holes. A company who has adopted the HFACS taxonomy should first classify its accident and near miss cases using the HFACS taxonomy. The company can then judge its findings against the benchmarking standards determined in Chapter Four. In order to enhance mitigations, association findings can help to identify other areas for mitigation or other areas, which may be affected by mitigation efforts.
Berry, Katherine, "A Meta-Analysis of Human Factors Analysis and Classification System Causal Factors: Establishing Benchmarking Standards and Human Error Latent Failure Pathway Associations in Various Domains" (2010). All Dissertations. 640.