r4 XEM: Leveraging AI/ML Predictive Analytics to Enhance Air Force Special Warfare Recruiting

The Active-duty Air Force is projected to miss its 2023 recruiting goal by 10%, amid a historic low unemployment rate and a growling lack of interest and eligibility to serve among young Americans. This is the first time the Air Force will miss its recruiting goals in twenty years. This challenge is even more pronounced for the Air Force Special Warfare (AFSPECWAR) community.  AFSPECWAR comprises ground combat forces that specialize in airpower application in hostile, denied and politically sensitive environments. AFSPECWAR requires Airmen with unique skills and abilities in order to perform highly specialized missions (often referred to as “unicorns”). The ideal candidate profile is similar to Navy SEALs, Special Forces, and Raiders as seen in the other military services.  Recruiting for these positions is inherently difficult. Current recruiting processes and socio-economic factors put AFRS at a disadvantage. First, AFRS recruiters must compete with the other military services’ special warfare recruiters. AFSPECWAR recruiters are also outnumbered by the other military services.  In low population density areas , this small number of recruiters must travel hours to meet and qualify candidates. As a result, recruiters face enormous difficulty meeting their recruiting targets. The consequences are clear. Simply put, the Air Force cannot fly, fight, and win without talented and motivated Airmen. To address this need, the Air Force Recruiting Service (AFRS) sees an opportunity to partner with r4 Technologies on a Small Business Innovation Research (SBIR) Phase II project to bring advanced  data science and AI/ML to its Air Force Special Warfare search and recruitment processes using r4’s XEM platform.  The 330th Recruiting Squadron (330 RS) seeks to partner with r4 Technologies to test and evaluate the r4 XEM Manpower Recruiting and Retention Platform (XEM MRR)  as a tool to micro-target geographic areas that carry the highest likelihood of AFSPECWAR service using predictive data analytics and geographic and demographic segmentation. The RDT&E of this project will modify and adapt the current XEM platform into a DAF-specific user interface that combines AFRS-provided recruitment data with publicly available information (PAI) to identify and target geographic areas with the highest likelihood of recruitment candidates, down to the neighborhood level. XEM MRR will analyze AFRS-provided and approved recruiting data, target metrics, and PAI to create modified/adapted “look-alike” profiles of ideal enlistment candidates, and then using its predictive analytics engine, provide recommendations to recruiters on where to target on the block (i.e., neighborhood) level.