Abstract
Self-assessment checklists (SACs) are used in the Air Force to evaluate the safety of the internal controls in each unit, ensuring they meet the required standards. However, a significant challenge arises with SACs due to the constant changes in regulations necessitating frequent updates to the checklist. Air Force workers, commanders, and other compliance officers spend roughly hundreds of hours a month going through Air Force regulation documentation, spanning over 74 pages. As a result, creating checklist questions based on these requirements can be extremely time-consuming when done manually. The objective of our research was to build an Artificial Intelligence-based tool to address this inefficiency. We did this by creating an algorithm, in Python, that first goes through a PDF, which in our case was an Air Force regulation document, and extracts sentences that contain the keywords "shall", "will", and "must". After extracting these sentences, we then normalized the text by replacing newline characters with spaces. Lastly, we split the normalized text into potential sentences. Then the algorithm can phrase these sentences into question form, to automatically generate questions. Going by the name of FlightList, our product aims to assist Air Force workers who have to manually create these SACs, allowing them to put their time and effort into less monotonous tasks that require more manpower. We hope this prototype can be integrated into the Air Force to make the process of turning regulations into checklists more efficient.
How to Cite:
, D., , S., , V., , E., , T., , M. & , K.,
(2024) “FlightList: Automating Self-Assessment Checklists for Optimal Compliance and Efficiency in Air Force Units”,
Journal of Student-Scientists' Research 6.
doi: https://doi.org/10.13021/jssr2024.4316
Files
Downloads are not available for this article.