The United States Army’s Automated Installation Entry (AIE) program is charged with the task of force protection via the application of specialized electronic security system integrating access control points, sensors, surveillance systems, biometric data, and monitoring capabilities for heightened access control. With such a critical task, system reliability is of upmost importance to the Army’s mission.
To solidify this facet, the U.S Army sought Boulevard’s cloud computing and data science capabilities to develop an architecture to streamline secure data sharing and facilitate real-time system performance assessments of AIE instances on geographically dispersed Army installations.
Key to Success and Results
To answer the call, Boulevard consultants developed an automated, cloud-hosted toolset, the Automated Installation Entry (AIE) Reliability Analytics Model (A-RAM) to enables site managers and program office personnel to extract real-time system performance and reliability data for over 200 continental united states (CONUS)-based installation security systems and report system faults to the program office for its existing suite of complex, large-scale personnel access systems.
Key tenets of this application (A-RAM) include 1) utilization of an adaptive machine learning algorithm rooted in multivariate regression analysis to facilitate the ability to forecast likely system failures and associated performance metrics to inform program office mitigation strategies and replacement part orders and 2) establishment of a secure and reliable system interface between A-RAM and the cloud-based performance data entity housing the secure AIE data.