Predicting Ranger Assessment and Selection Program 1 Success and Optimizing Class Composition
Abstract
The 75th Ranger Regiment is a US Army Special Operations unit responsible for executing raids and forcible entry missions across the globe within 18 hours of notification. In this thesis, we conduct the first data analysis and optimization of Ranger Assessment and Selection Program 1 (RASP1). RASP1 is an eight-week selection for volunteers in the grade of E1 (Private) to E5 (Sergeant) implemented up to ten times per year. We create logistic regression and partition tree models to identify significant factors that contribute to a candidates success at RASP1 and predict graduation rates. We use an integer linear program (ILP) to prescribe the number of soldiers by grade and Military Occupational Specialty to bring to each RASP1 class to efficiently fill required billets across all units in the Ranger Regiment. We provide the Ranger Regiment leadership with flexible models that offer insight to support their manning decisions. We show effects on RASP1 class composition with changes to capacity constraints, input parameters, and demand. For example, we find the Ranger Regiment could reduce the number of annual RASP1 classes from ten to eight based on several realistic assumptions. Such an annual reduction could save hundreds of man hours and significantly reduce training resource requirements (e.g., ammunition, land use, barracks and food). We encourage detailed exploration of our underlying assumptions and continued use of the ILP.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jun 01, 2017
- Accession Number
- AD1046554
Entities
People
- Anthony D. Smith
Organizations
- Naval Postgraduate School