A human error classification system for small air cargo operators

The Aviation industry is a highly complex and dynamic domain. Over all, commercial flight crews consistently operate at a very high level of reliability and safety. Yet, accident records show there is a disparity between the flight crews that operate under Title 14, Part 121 of the Code of Federal Regulations (CFR) and those that operate under Part 135 of that same code. In their daily operations, the performance of both groups are shaped by the complexity of this environment, their interactions with the system and their own personal, as well as team skill sets. However, the flight crews of part 135 operators consistently make more errors, ranging from procedural, tactical and regulatory. These factors have been studied from a broad theoretical framework using many different perspectives, but a conclusive explanation for the disparity in the accident rates between the part 121 and 135 operators remains elusive. One common methodology of error classification is analyzing a database of accident and incident information to identify the errors that pilots make in specific operational areas within the aviation system. In the last decade, researchers have developed a number of error classification schemes, and the reports of their findings are abundant in the literature describing the taxonomy of human errors in the aviation system. However, there is little research that correlates the flight training methodology that is designed to mitigate these errors, to the error classification schemes commercial air carriers currently use. Furthermore, there is no research that focuses on the classification errors made by pilots or flight crews that operate under the part 135 regulations. This thesis examines some of the most influential literature that has shaped the development of systems designed to analyze and encode aviation accidents and incidents, as well as systems to classify human error in the aviation system. This thesis examines the structure and elements necessary to develop an effective human error classification system, the methodology used to design classification systems in general, as well as the taxonomy used to develop human error classification systems. This thesis reviews the methodology used in the current aviation human error classification systems. Additionally, it proposes a preliminary model for a system designed to classify pilot and flight crew error that occurs during the operation of commercial aircraft under part 135 regulations, as well as suggests corrective actions to mitigate these errors. This system is based on the development of a theoretical concept for identifying, analyzing, encoding, and classifying flight crew error. It lists corrective actions in a terminology that can be used to develop flight-training activities and scenarios that will reduce the number of errors pilots and flight crews make during part 135 regulated flight operations. This thesis reports the analysis of trials, in which part 135 flight instructors and or check airman, as well as flight instructors that are licensed and regulated under part 91 of the Federal Aviation Regulations classified five randomly selected reports from the Aviation Safety Reporting System. A comparison of the results of the classifications made by part 91 instructors, verses the part 135 instructors and or check airman will be discussed. Finally, based on the finding of the analysis of these trials, recommendations for improvements in the design and implementation of future error classification systems designed to mitigate the pilot errors made during commercial flight operations are discussed.