Optimal scheduling of Army initial entry training courses

Abstract

Scheduling Army enlisted initial entry training is a complicated task currently done manually at the U.S. Army Training and Doctrine Command Headquarters, Fort Monroe, Virginia. Scheduling results are entered into the Army's automated training system used by both training centers and recruiters to assign eulistees to training spaces at training centers. This thesis develops a mixed integer program to plan monthly training schedules for Basic Combat Training, One Station Unit Training, and Advanced Individual Training. The goals are to maximize the efficiency of the training schedule (by minimizing the number of recruits held over), to minimize the annual soldier training requirements not met, and to aspire to optimally fill courses. The model is implemented in the CAMS modeling language. The output is a matrix of 230 courses to 50 assigned start weeks. This approach accomplishes 94 percent of the annual Army requirements for fiscal year 2000 (FY00). Holdover time is decreased to 90, 360 soldier-weeks using the optimal scheduling method compared with 180,000 weeks projected for FY00 using existing methods. This improvement saves 1800 soldier-years, or a brigade's worth of manpower for the Army at no additional cost. This approach effectively creates over 5500 additional training seats. This model should be implemented as a methodology for scheduling Initial Entry Training courses.