What Happened, and Why: Toward an Understanding of Human Error Based on Automated Analyses of Incident Reports

The objective of the Aviation System Monitoring and Modeling (ASMM) project of NASA s Aviation Safety and Security Program was to develop technologies that will enable proactive management of safety risk, which entails identifying the precursor events and conditions that foreshadow most accidents. This presents a particular challenge in the aviation system where people are key components and human error is frequently cited as a major contributing factor or cause of incidents and accidents. In the aviation world, information about what happened can be extracted from quantitative data sources, but the experiential account of the incident reporter is the best available source of information about why an incident happened. This report describes a conceptual model and an approach to automated analyses of textual data sources for the subjective perspective of the reporter of the incident to aid in understanding why an incident occurred. It explores a first-generation process for routinely searching large databases of textual reports of aviation incident or accidents, and reliably analyzing them for causal factors of human behavior (the why of an incident). We have defined a generic structure of information that is postulated to be a sound basis for defining similarities between aviation incidents. Based on this structure, we have introduced the simplifying structure, which we call the Scenario as a pragmatic guide for identifying similarities of what happened based on the objective parameters that define the Context and the Outcome of a Scenario. We believe that it will be possible to design an automated analysis process guided by the structure of the Scenario that will aid aviation-safety experts to understand the systemic issues that are conducive to human error.