Degradation Analysis of Gun Barrel Life Using Machine Learning and Design of Experiments

Abstract

>The US Army M240 Small-Caliber Machine Gun Barrel degrades as round-count increases, leading to velocity drop and mismatch between aiming/fire control and round impact at range. This degradation can be understood as a functional response of velocity change over time, with time in this case represented by round-count on the barrel. Statistically modeling this type of functional response poses some additional challenges when compared to traditional one-dimensional data, and Functional Data Analysis (FDA) can help address this added complexity. Although FDA is not a new technique, its application to gun barrel degradation had never been employed before by the Department of Defense (DoD) until now.