Project Abstract:

Aviation Safety Action Program

• Industry: Aviation
• Business Need: Human Error Analysis
• Client: Federal Aviation Administration

The recent introduction of the FAA’s voluntary Aviation Safety Action Program (ASAP) has provided carriers with a powerful tool for understanding the human factors issues that their crews face during typical line operations. Unfortunately, this tool has rarely been used to its fullest potential. We believe that this is because most ASAPs collect human factors data using text narratives that require costly and time-consuming content analysis. To address this problem, we are developing a human factors taxonomy and electronic data reporting/analysis tool for use in ASAP. When complete, these products will be provided at no cost to industry.

This project has two main goals. The first is to develop a taxonomy for classifying human factors issues in de-identified ASAP reports. The second is to embed this taxonomy within a searchable computer database. The end product will be a database that can help airlines proactively identify threats to safety. This information generated from these databases can then be used for a variety of reasons, such as identifying training objectives or developing Line-Oriented Flight Training (LOFT) scenarios.

We conducted an in-depth review of existing aviation human factors taxonomies, accident/incident reporting systems, and data reporting tools. A test-bed of over 300 human factors issues was developed from existing ASAP reports, theoretical models of human error, and human factors textbooks. This list of 300 factors was culled to 91 by eliminating redundant factors. Next, 23 airline pilots from a regional air carrier sorted the factors into groups using a web-based sorting program. The results were used to develop seven causal categories. Next, approximately 100 pilots from the same airline participated in a web-based survey; pilots were asked to assign each of the 91 factors to one of the seven categories. Items that were not consistently sorted were reconciled by senior level SMEs. The resulting taxonomy was tested by having ASAP meeting attendees select the causal category and factor for de-identified ASAP reports. An acceptable level of agreement was reached, thus finalizing the taxonomy. The taxonomy is currently being implemented by the FAA and its contractors into the web-based ASAP software program.