A Review of Selected Aviation Human Factors Taxonomies, Accident/Incident Reporting Systems, and Data Reporting Tools
The Aviation Safety Action Program (ASAP), which collects de-identified incident reports from line pilots regarding threats to safety (Federal Aviation Administration, 2000), is very similar to NASA’s Aviation Safety Reporting System (ASRS; Federal Aviation Administration, 1987). For example, both systems encourage pilots to self-report their errors by providing limited immunity from prosecution. However, unlike the ASRS, the ASAP is carrier-specific. Drawing on the ASRS example, many ASAPs collect background information regarding the pilot who submits the report and the flight conditions that immediately preceded the event. In addition, space is usually provided so that the reporting pilot can write a short narrative describing the event, the causal factors that precipitated it, and suggestions for preventing its reoccurrence. As a general rule, most of the Human Factors issues described in ASAP reports can be found within these narratives.
In a perfect world, voluntarily-collected safety data such as ASAP incident reports, line audit summaries, and Flight Operations Quality Assurance (FOQA) output would be used to continuously assess carrier safety and to suggest recommendations for improvement. Unlike line audits and FOQA output that are designed to answer the question “What happened?”, ASAP reports are uniquely designed to answer the question “Why did this event happen?”. Unfortunately, because the analysis of textual data is extremely difficult, ASAP data has rarely been used to its fullest capacity, such as in developing training objectives, Line Operational Simulation (LOS) scenarios, and other safety-related interventions (for a notable exception, see Baker, Chidester, & Brannick, 1998). To remedy this problem, we are developing a generalized Human Factors taxonomy and electronic reporting/analysis tool that will be made freely available to industry. This tool will use pull-down menus and check-in-the-box items to supplement the text narrative, thereby allowing carriers to easily classify and quantify the root causes of pilot error. Eventually, this electronic reporting tool will include sophisticated data analysis and graphing tools to help carriers identify the most pressing problems, develop data- driven interventions, and assess the effectiveness of these interventions by statistically comparing pre- and post-intervention data. This review—which summarizes the state-of-the art in aviation error reporting, classification, and analysis—serves as the foundation for our future taxonomic research.