Using Monte Carlo simulations to translate military and law enforcement training results to operational metrics. (July 2022)

• Record Type: Journal Article
• Authors: Biggs, Adam T.; Hirsch, Dale A.
• Is Part Of: Journal of defense modeling and simulation. Volume 19:Number 3(2022)
• Page Start: 403
• Page End: 415
• Publication Date: 2022-07
• Subjects: Military -- law enforcement -- Monte Carlo -- translational -- applied research; Military art and science -- Computer simulation -- Periodicals; 355.0011305
• DOI: 10.1177/15485129211021159
• Languages: English
• ISSNs: 1548-5129

Abstract

There are numerous challenges comparing research initiatives due to methodological differences and scenario-specific problems. Military and law enforcement issues present an extreme variant of this challenge. Specifically, assessment and training scenarios strive for realism, but operators cannot engage one another with live rounds or induce the full spectrum of environmental stressors for obvious safety reasons. Instead, particular factors are evaluated in a given scenario via experimental statistics despite the inherent difficulty in communicating inferential statistics to the intended audience of military and law enforcement professionals. The current investigation explores how Monte Carlo simulations can use probabilistic distribution sampling to convert statistical inferences into concrete operational outcomes. Using this type of distribution sampling, statistical inferences can be translated into operational metrics such as the probability of winning a gunfight. Describing these statistical values and effect sizes in terms of survival provides a more appreciable operational metric that military and law enforcement personnel can use when evaluating the advantages of various training platforms or equipment. Several approaches are examined that each accomplish this general goal, including circumstances outside of marksmanship and lethal force decision-making.