The pair were part of a three- to six-month fellowship program involving 40 students from 10 U.S. universities. They get a chance to work with one of 12 military units.
(Other schools include Duke and the universities of North Carolina, Colorado at Boulder, Florida Polytechnic University and Washington University in St. Louis.)
The UCSD duo joined data scientists from Commander, Naval Air Forces, at North Island Naval Air Station for what the Navy calls the first-ever collaboration between the National Security Innovation Network’s (NSIN) X-Force program and CNAF.
The Army, Marine Corps and Air Force also take part in the X-Force program.
With their assigned unit, the students used their skills in areas like mobile and web app development, social media strategy and data analysis to solve a real-world problem unique to each individual military unit.
CNAF’s Force Readiness Analytics Group (FRAG) analyzes data in the field of aviation readiness, using the information to make informed, data-driven decisions.
The FRAG primarily aims to reduce the time an aircraft is non-mission capable by utilizing data to optimize maintenance scheduling and maximize performance.
“We stood up the FRAG in September 2018,” said Senior Chief Machinist’s Mate Aaron Thiss, a member of the CNAP FRAG Team. “Our maintainers were already doing all the maintenance they could, but with so much demand on them it was hard to figure out which aircraft to prioritize.
“That’s where we came in. We started analyzing data we already had to streamline the maintenance process, which will have major positive effects down the line on readiness, maintainer experience, pilot experience and more.”
NSIN started the X-Force fellowship program this summer and sent three students to the FRAG to help develop a statistical model to predict when an F/A-18E Super Hornet would need preventative maintenance.
By analyzing data and utilizing machine learning (the scientific study of algorithms and statistical models), the students and the FRAG team work together to better maintain aircraft by creating a prioritized list of which aircraft need maintenance first, saving time, money and manpower, as well as potentially avoiding in-flight incidents.
“FRAG was tasked with trying to use analytics to fix the low number of mission ready jets, so we are developing a machine-learning model to predict when these jets will break,” said Jordan Coursey, a recent international affairs masters graduate and X-Force fellow at UCSD.
“So we can either schedule maintenance to prevent breakage or give [maintenance] control some sort of hierarchy to choose which planes are more or less likely to break,” said Coursey, who once played roller hockey for UCSD.
Thiss said: “We have to man, train and equip the fleet, it’s important that we really understand the problems in those areas and then figure out how to provide those solutions. So by analyzing this data and becoming more proficient at it, then we can spur other areas and other organizations through the fleet to do the same.”
As the program draws to a close on its first year, the students have been able to reflect on their experiences and what they hope to see their work with X-Force will become going forward.
“I think servicing the U.S. armed forces really benefits the world in a lot of ways, so I’m really excited to be able to do that,” said Aj Langley, a computer sciences masters student and X-Force fellow at UCSD.
“I hope the program keeps going and I hope that as a result of this and the FRAG’s efforts that the military is able to embrace predictive analytics and machine learning even more.”
Coursey said that the difference between operations in the private sector and the military are interesting to experience firsthand, but he hopes to one day be part of bringing the two sectors closer together.
“Personally, I have learned so much from this program, working in such a large organization, working with several different clients — those are all things that have been very beneficial to me.”
>> Subscribe to Times of San Diego’s free daily email newsletter! Click hereFollow Us: