Aviation safety reports 1,2 are the best available source of information explaining why a flight incident happened. However, stream of consciousness permeates the narratives making the automation of the information extraction task difficult. We propose an approach and infrastructure based on a common pattern specification language to capture relevant information via normalized template expression matching in context. Template expression matching handles variants of multi-word expressions. Normalization improves the likelihood of correct hits by standardizing and cleaning the vocabulary used in narratives. Checking for the presence of negative modifiers in the proximity of a potential hit reduces the chance of false hits. We present the above approach in the context of a specific application that is the extraction of human performance factors from NASA ASRS reports. While knowledge infusion from experts plays a critical role during the learning phase, early results show that in a production mode, the automated process provides information that is consistent with analyses by human subjects.

The purpose of this study was to analyze the frequency of general aviation (GA) airplane accidents and accident rates on the basis of aircraft certification to determine whether or not differences in aircraft certification rules had an influence on accidents. In addition, the narrative cause descriptions contained within the accident reports were analyzed to determine whether there were differences in the qualitative data for the different certification categories. The certification categories examined were: Federal Aviation Regulations Part 23 (Part 23), Civil Air Regulations 3 (CAR 3), Light Sport Aircraft (LSA), and Experimental-Amateur Built (E-AB). The accident causes examined were those classified as: Loss of Control (LOC), Controlled Flight into Terrain (CFIT), Engine Failure, and Structural Failure. Airworthiness certification categories represent a wide diversity of government oversight. Part 23 rules have evolved from the initial set of simpler design standards and have progressed into a comprehensive and strict set of rules to address the safety issues of the more complex airplanes within the category. E-AB airplanes have the least amount of government oversight and are the fastest-growing segment. The LSA category is a more recent certification category that utilizes consensus standards in the approval process. CAR 3 airplanes were designed and manufactured under simpler rules, but modifying these airplanes has become lengthy and expensive. The study was conducted using a mixed-methods methodology. A Chi-Square test was used for a quantitative analysis of the accident frequency among aircraft certification categories. Accident rate analysis of the accidents among aircraft certification categories involved an This article is based on the Doctoral Dissertation of Carolina L. Anderson, submitted to the Department of Doctoral Studies in partial fulfillment of the requirements for the degree of Doctor of Philosophy at Embry-Riddle Aeronautical University.

No abstract

No abstract

No abstract