Date of Award

12-2009

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Legacy Department

Industrial Engineering

Advisor

Shappell, Scott A

Committee Member

Duchowski , Andrew

Committee Member

Ferrell , William

Committee Member

Figliola , Richard

Committee Member

Kimbler , Delbert

Abstract

Over the last 20 years the number of annual accidents in general aviation has seen a steady decline. The annual number of fatal accidents, however, has remained relatively stable. Adverse weather is consistently cited as a leading cause of fatal accidents in general aviation; weather accidents have a fatality rate nearly four times that of other general aviation accidents. Reviews of accident data reveal that often these accidents occur when inexperienced pilots who are certified only for visual flight rules flight continue into adverse weather. The Situation Assessment Hypothesis suggests these accidents occur when pilots fail to accurately recognize weather conditions they are facing. Cue-based training has been suggested as an effective method for improving situation assessment. A detailed analysis of the ways cue-based training affects decision-making behavior is needed to better understand the training's effectiveness.
This research aims to first analyze the way pilots of varying experience levels view weather scenes and make weather-related decisions. Second, it endeavors to quantify the effects of cue-based training on weather-related decision making both in terms of decision accuracy and visual scan behavior. Beyond weather-related decision making, this study provides a methodology for evaluating experience and training that could be extended to any type of primarily visual task.
Findings of the research indicated clear differences between pilots and non-pilots with minor differences seen between pilots of different experience levels. Non-pilot scanpaths consisted of many shorter fixations spread over a longer period of time compared to pilots' scanpaths. High-time pilots made their first fixation more quickly while directing fewer fixations to areas of cloud darkness than low-time pilots.
While the training program failed to improve decision accuracy, it successfully affected a more conservative bias in groups, thereby reducing the number of simulated flights into adverse weather. What's more, the training reduced the amount of visual information needed to make a decision in all groups, with largest effects seen among non-pilots. This suggests the training program was effective and should be implemented early in the pilots' career in order to achieve the maximum benefit.

Share

COinS