Date of Award

5-2026

Document Type

Dissertation

Degree Name

Doctor of Aviation (AvD)

Department

Aeronautics

First Advisor

Brook E. Wheeler

Second Advisor

William W. Arrasmith

Third Advisor

John E. Deaton

Fourth Advisor

Gregory A.D. Fox

Abstract

This study examined whether the elements that comprise an Instrument Flight Procedure (IFP) could predict when a procedure had more Aviation Safety Reporting System (ASRS) records. Previous studies focused on aircrew performance related to procedure charting and lacked automated methods for selecting data. This correlational, predictive design used a census of IFP procedures in U.S. airspace and ASRS records between 1 January 2013 and 31 December 2023, and researcher-created software to evaluate the data. Software reliability was examined using an Intra-class Correlation Coefficient (ICC) of randomly selected procedure data evaluated by two pilots and three aeronautical data Subject Matter Experts (SMEs). The single-measures ICC of .842 (p < .001) indicated the software was reliable for measuring the Procedure Task Load (PTL). The average measures ICC of .970 (p < .001) indicated extremely high agreement between the SMEs and the software. Software searched the ASRS records for the names of IFPs. The reliability of the search software was assessed by having SMEs review 80 randomly selected ASRS records and answer a dichotomous question of agreement with the search results. Cohen’s Kappa was used to measure the two raters’ evaluations, which showed high agreement (κ = .739, p < .001). The average PTL was obtained by dividing the PTL by the total distance covered by a procedure. ASRS records were averaged by dividing the number of Aeronautical Information Regulation and Control cycles the procedure existed during the study period. A linear relationship was not established because the assumptions were not met. A supplemental analysis revealed that the median number of altitude restrictions in approach procedures for the high complexity group was significantly larger (MdnHigh = 10.91, MdnLow = 7.48, U = 2,236,825.5, z = -99.178, p < .001, r = .801), compared to the low complexity group when approaches were segregated by PTL (low group PTL 5–25, n = 8,537, high group PTL 26–162, n = 6,806). The software automated evaluation, and although a linear relationship predicting the ASRS frequency based on PTL or APTL was not discovered, the supplemental analysis revealed significant differences in safety incidents by procedure complexity.

Download

Available for download on Monday, November 09, 2026

Share

COinS