Toward a Generalized Human Factors Taxonomy for Classifying ASAP Incident Reports, AQP Performance Ratings, and FOQA Output
In R.S Jensen (Ed.) Proceedings of the 12th International Symposium on Aviation Psychology
Over the years, the FAA has partnered with industry to develop programs for reporting, classifying, and analyzing safety-related data, but none has been able to integrate data from multiple sources. We are developing a generalized Human Factors taxonomy for classifying de-identified ASAP incident reports, AQP performance ratings, and FOQA output. Eventually this taxonomy will be embedded into a series of searchable computer databases that speak a common language, allowing the search for trends.
Our project has three primary goals. The first goal is to develop a comprehensive ASAP taxonomy that will allow carriers to quantify the Human Factors issues that their crews face during typical line operations. Armed with this information, carriers can then develop data-driven interventions and evaluate the effectiveness of these interventions by statistically comparing pre- and post-intervention data.
Our second goal is to embed this taxonomy within a searchable data collection and reporting tool. Doing so will streamline the process of collecting, managing, and reporting ASAP data. As a result, the carriers' limited resources can be devoted to more goal-directed tasks such as problem identification, analysis, and resolution.
Our third goal is to extend all or part of this taxonomy to include de-identified AQP performance ratings and de-identified FOQA output. The end result will be three separate databases that use a common Human Factors taxonomy. Although individual records will be de-identified, carriers will be able to identify safety-related problems by triangulation. For example, if a carriers' ASAP reports indicate that non-precision approaches are a problem, their FOQA and AQP data can be analyzed to verify the problem's existence.
The value-added benefits of this project include: the capacity to identify problems by triangulation, the capacity to rank order safety-related problems by frequency of occurrence or perceived probability of reoccurrence, the capacity to develop data-driven interventions, and the capacity to measure the effectiveness of these interventions by statistically comparing pre- and post-intervention data.