Using experimental design and data analysis to study the enlisted specialty model for the U.S. Army G1

Abstract

Every month the U.S. Army G1 uses an Enlisted Specialty (ES) model consisting of a simulation and an optimization to forecast the Army's enlisted manpower program by Military Occupational Specialty and grade. The model is responsible for operating a 30.64 billion dollar manpower program that currently manages 460,000 enlisted Soldiers. The research in this thesis studies the objective function coefficients associated with decision variables in the ES optimization model. Experimental design and analysis techniques were used to study how changes in the coefficients affect the assignment of current enlisted soldiers to vacant positions in the Army. Results of the thesis show that by adjusting eight of the coefficients in the optimization model, the deviation between authorizations and inventory can be reduced by 14%. This improves the U.S. Army's force structure alignment and ensures the Army is ready to fight the nation's wars.