A proactive safety enhancement methodology for general aviation using a synthesis of aircraft performance models and flight data analysis

Abstract

As general aviation (GA) industry and its operations have grown along with the aviation industry development, improving aircraft safety has been a key interest in the GA industry. According to the U.S. Department of Transportation, GA in the U.S. has been suffering higher fatal accident rate compared to that of scheduled airline flights. This statistic indicates that safety enhancement effort is inevitable and reduction of GA aircraft fatality rate needs to be a prioritized goal in the GA community. The increasing pervasiveness of data-driven-safety programs such as flight data monitoring (FDM) in commercial aviation has permeated GA, giving rise to a growing body of quantitative safety analysis opportunities. FDM and other data-driven programs such as flight operations quality assurance (FOQA) feature a retrospective analysis of flight data records that identify potential safety-critical phenomena and the formulation and implementation of corrective actions. Thus, quantitative aircraft performance modeling and simulation capabilities emerge as critical enablers for safety analysis, particularly when coupled with flight data records that produce a rich and meaningful picture of operational safety. However, the intended application of the operational safety analysis imposes essential requirements on GA aircraft models and flight data records to be used by safety analysts. First, models must provide predictive capabilities with high flexibility and accuracy over the wide range of operational conditions. Also, to maximize the benefits of data-driven safety analysis, securing tidy data that is ready to be analyzed is as important as the ongoing collection and analysis of flight data records. Thus, the objective of this study is to develop a proactive operational safety analysis method by introducing a realistic and flexible performance modeling method and an efficient data noise removal technique for a fixed-wing GA aircraft. To accomplish the research goal, this study explores various existing performance modeling methods to propose a more cost-effective data-driven aerodynamic model that can adequately predict aircraft performance and capture the unsafe aerodynamic behavior of a fixed-wing GA aircraft. Furthermore, this study examines various data noise filtering techniques in both time and frequency domains to suggest an affordable and effective data cleaning process while preserving true aircraft behavior. Finally, this research suggests a quantification methodology for an operational safety assessment of GA fixed-wing aircraft using the previously obtained accurate and affordable aerodynamic model and clean flight data. The suggested safety assessment procedure enables a better understanding of realistic aircraft performance by adding flexibility to identifying operational limits and ensuring the reliability of collected flight data.