Impact of Human Characteristics in Basic Underwater Demolition/Seal (BUD/S) Training Performance

Abstract

Reporting and clerical functions at Naval Special Warfare (NSW) Echelon IV commands are ripe fordigitization, automation, and optimization. This study utilizes a restricted digitalized NSW dataset to showcase how “big data” in the context of SEAL training can be used to predict performance success of various Basic Underwater Demolition/SEAL (BUD/S) training evolutions. Our study focuses on multiple human characteristics and compares their correlation to evolution pass rates in training using Ordinary Least Squares (OLS) for our prediction model. From our initial regression analysis of over 232,000 data points, our findings indicate higher pass rates for BUD/S candidates who are older, married, and officers, as well as increased pass rates in individuals who were taller, lighter, and right-handed. Lower pass rates are found among minorities. The Black population had high fail rates in the evolutions that involve water activities. This study is an example of how long-term efficiencies could be gained from greater automation of data using simple software that could provide long-term benefit if captured in a more persistent and accurate manner. We advocate for the implementation of a more automated data/software collection system that can capture each student's training career in one cohesive data profile. Moving forward, NSW studies should continue to leverage the use of “big data” to optimize its performance across all domains of the force.