Photo Credit: Army National Training Center and Michael Crews
The US Army has always had a reputation for discipline, order and adherence to its systems. So who did it turn to when it needed to improve its training methods? Industrial Engineers, of course. As part of the Research Experience for Undergraduates (REU) Program, Nicholas Leone and Jack Werner joined grad student and Army Captain Brandon Lawrence to optimize the Army National Training Center’s (NTC) After-Action Review (AAR) process.
Forming the Team
How did this trio come together? ISE professor Brandon McConnell was the lynchpin to forming the team. For Werner, finding the research position “was as easy as walking through 111 Lampe Hall and knocking on doors.” He came across McConnell while going door to door to see what opportunities were available.
Leone heard about the opportunity in class. “I was instantly intrigued by the projects and type of work in the REU program,” he said and wanted the chance to help. He was advised to talk to McConnell about the research project, and McConnell happened to be Lawrence’s thesis advisor.
Lawrence was looking for some extra hands to help him comb through the large amounts of data and was eager to have additional assistance. “Jack and Nicolas understood the direction I wanted to go with the project and anticipated many of the questions that I had for them,” Lawrence recalled. With Werner and Leone’s quick thinking and Lawrence’s perseverance, this quickly became the perfect group to solve the problem at hand.
About every thirty days, a different Army unit arrives at the NTC to train in a tough and realistic environment. One part of this experience is the After-Action Review (AAR), a periodic assessment of different troop formations like brigade, battalion, and platoon. The AAR tracks the unit’s performance throughout their rotation. Lawrence’s goal was to quantify the rotational units’ effectiveness.
To determine each units’ effectiveness, Werner helped compare the data to streamline each soldier’s performance. His skill in coding and data management grew as he worked on the project. “On day one, I was limited to basic Excel,” Werner stated. “Now, three semesters later, I am highly proficient in Python, R, data cleaning, and data visualization.”
At the same time, Leone helped a team of students develop a dashboard that displayed the results from a laser tag tank battle. Leone said, “We built a map that shows where the tanks are and how well each team: friendly and enemy; are doing based on Army principles.” Having never coded with R before, Leone learned how to create metrics and analytics as well as how to use a program, Shiny, to create dashboards.
Lawrence developed the measures of performance based on the Army doctrine. “After measuring the unit across all measures of performance throughout their rotation, I seek to determine how well these factors predict operational effectiveness,” he explained. The goal is to use this information during AAR to evaluate the different teams and see how effective they were during training.
The project wasn’t a walk in the park, however. “The biggest challenge has been the sheer quantity of data that exists,” Lawrence explained. “For each individual training rotation, there is a database which consists of hundreds of tables filled with millions of rows of data that require complex queries to make discernable results.” Even after selecting the most useful data, there is still a large amount of data cleaning required to draw meaningful findings.
Another issue arose as Leone struggled with the syntax of R. “Many of the problems that I faced, especially in developing the dashboard, were very simple syntax mistakes that took me a long time to correct,” he recalls. Also, conducting research was new to Leone, and he had to learn the differences in documenting and presenting findings for a project like this one. Alongside these issues was the larger problem of COVID-19. “Obviously, the pandemic was a huge unforeseen circumstance last spring,” said Werner. “However, the transition to Zoom research meetings went quite smoothly, and our team definitely succeeded.”
Despite the challenges this project presented, the team did succeed and this will allow the NTC to provide more data-driven AARs. “By having my work running real-time during a training rotation, units will be able to see how their measures of performance influence their overall operational effectiveness,” explained Lawrence. The work could also be scalable for different troops and could be tailorable to allow the unit commander to focus on specific areas. “I would just like to thank Jack and Nicolas for their hard work throughout the semester,” Lawrence said. “Their efforts were invaluable.” And, of course, “I would also like to thank Dr. McConnell for providing me with this great resource.”