Date of Award
Master of Science (MS)
Stringfellow , Paris
Shappell , Scott
The Human Factors Analysis and Classification System (HFACS) is used to help identify causal factors that lead to human error. The framework has been used in a variety of industries to identify leading contributing factors of unsafe events, such as accidents, incidents and near misses. While traditional application of the HFACS framework to safety outcomes has allowed evaluators to identify leading causal issues based on frequency, little has been done to gain a more comprehensive view of the system's total risk. This work utilizes the concept of event severity along with the HFACS framework to help better identify target areas for intervention among unsafe events in wind turbine maintenance.
The objective of this work was to determine if there are any relationships between the certain HFACS causal factors and an incident's severity. The analysis was based on 405 cases which were coded for contributing factors using HFACS and were rated for actual and potential severity using a 10-point severity scale. Models for predicting potential and actual severity were generated using logistic regression. These models were then validated using actual data. Although the findings were not significant, it was determined that decision errors and preconditions to unsafe acts: technological environment were major contributors to events with high potential severity.
One limitation of this work was the limited availability of complete data on which to conduct the analysis. So, while the analysis produced non-significance, it is anticipated that as more data becomes available, the models will yield more concrete findings. Regardless, understanding the relationships among incident causal factors and outcomes may shed light on those causal factors which have the potential to lead to catastrophic events and those which may lead to less severe events.
Kavade, Hrishikesh, "A LOGISTIC REGRESSION MODEL TO PREDICT INCIDENT SEVERITY USING THE HUMAN FACTORS ANALYSIS AND CLASSIFICATION SYSTEM" (2009). All Theses. 712.