The security environment is saturated by technological advancements, creating an imperative to increase the speed with which the Army Sustainment Enterprise generates and maintains combat power as the world enters great power competition. To meet these threats, the Army is advancing its predictive logistics initiative. Army Forces Command (FORSCOM), supported by Army Materiel Command (AMC), Army Futures Command, Training and Doctrine Command (TRADOC), and the Assistant Secretary of the Army (Acquisition, Logistics, and Technology) (ASA (ALT)), leverages technological advancements in artificial intelligence and predictive analytics to inform advancements to better operate as part of the joint force to compete, penetrate, disintegrate, and exploit adversaries in a multidomain environment.
The Spartan Brigade, 2nd Armored Brigade Combat Team (2ABCT), 3rd Infantry Division (3ID) at Fort Stewart, Georgia, was selected by FORSCOM to demonstrate predictive maintenance to observe, evaluate, and inform requirements for Enterprise Business System — Convergence and platform capability development documents. Central to the demonstration is Soldier feedback in the operational force driving the observations and informing the generating force. This approach allows the generating force to speed up the acquisition process.
Due to the breadth and depth of predictive logistics coverage, the Army takes a pragmatic approach to demonstrate the end-to-end process by leveraging various existing Army equipment and experimenting with repurposing equipment in new ways. In this process, observations are documented on the strengths and weaknesses of interfacing equipment. Positive outcomes have been realized because of repurposing existing equipment, including potential deep cost savings, as well as improved warfighting and business processes. Thus far, the demonstration includes garrison operations and a National Training Center rotation at Fort Irwin, California, to obtain insights from the complexity of a field environment.
AMC tasked the U.S. Army Tank-automotive and Armaments Command to facilitate the end-to-end demonstration within 2ABCT, 3ID in concert with ASA (ALT) program management (PM) offices and the TRADOC Sustainment Center of Excellence. The brigade demonstration enables the discovery process to capture observations for lessons learned. These consolidated observations are directly delivered to authors of requirements documents, informing them of requirements for ASA (ALT) PMs. The Program Executive Office Command, Control, Com-munications-Tactical (PEO C3T) was selected by ASA (ALT) as the office of primary responsibility for materiel development for predictive logistics and predictive maintenance. PEO C3T is actively engaging and coordinating with platform and communication PMs to synchronize effects to deliver the Army of 2030 and design the Army of 2040.
Predictive maintenance is a maintenance strategy that uses data and analytics as well as artificial intelligence and machine learning (AI/ML) algorithms to predict when equipment is likely to fail so maintenance can be performed proactively before the failure occurs. This approach can significantly improve equipment availability, reduce maintenance costs, and extend the lifespan of the equipment. Predictive maintenance is a sub-component of the Army’s predictive logistics strategy and Class III (fuel) and Class V (ammunition) distribution.
The predictive maintenance process involves five key stages of the end-to-end data flow process: collect, distribute, store, analyze, and visualize. All five stages are wrapped by military-grade encryption, enabling a zero-trust architecture to protect the network and the data. These stages work together to turn data from equipment sensors into actionable insights that can be used to improve the Army’s equipment management.
The first stage of the process is collecting data from sensors placed on the vehicles as well as operators’ preventative maintenance checks and services (PMCS) process. The sensors gather data on various aspects of the equipment’s performance, such as engine temperature, oil pressure, and fuel consumption, and the operator PMCS tablets capture the faults that aren’t discoverable by a system sensor.
The second stage of the process is distributing data via a transport mechanism such as satellite communication or cellular networks to get the data to a centralized repository. This enables the data to be accessed and analyzed by AI/ML algorithms that make up the next stage of the process. The use of Joint Technical Data Integration (JTDI) to synchronize data across tactical and enterprise nodes ensures the correct data is sent where necessary with minimal consumption of valuable bandwidth. JTDI is a Navy program of joint interest that supports the interoperability required for multidomain operations.
The third stage is storing the data in a centralized repository. This enables the data to be organized and analyzed, helping identify patterns and trends. Storage and analysis of data are intended to occur at multiple echelons, from the tactical to the strategic level. Governance of data is critical throughout this data distribution process.
The fourth stage of the process is analyzing the data using machine learning algorithms. This is where the predictive maintenance magic happens. The algorithms look for patterns and anomalies in the data, using the information to predict when equipment will likely fail.
The final stage of the process is visualizing the data, making it easier to understand and act upon. The Army uses dynamic and multilayered dashboards and visualizations to monitor the performance of the equipment, identify potential issues, and schedule maintenance.
Predictive maintenance has many benefits for the Army, including:
- Increased equipment availability: By predicting when equipment is likely to fail, the Army can perform maintenance proactively, keeping its equipment in top condition and increasing its availability.
- Reduced maintenance costs: Proactive maintenance reduces the costs associated with unscheduled maintenance, such as repairs, replacement parts, and labor.
- Improved equipment lifespan: Predictive maintenance helps extend the lifespan of the Army’s equipment by identifying potential issues before they become major problems.
The Army’s experimentation with predictive maintenance is a major step forward in its equipment management strategy. The current demonstration efforts are focused on maintenance data automation as a key building block to start the foundation for maneuver and sustainment mission command planning. Mission analysis is underway to integrate data flows from maintenance, ammunition, and fuel distribution business processes into mission command data pathways.
Senior leaders are engaged with stakeholders from the tactical to strategic levels to drive implementation for the Army. The end-to-end data flow process of collecting, distributing, storing, analyzing, and visualizing transforms data at the tactical level into actionable insights for all echelons, posturing the Army to be responsive where needed.
Benjamin D. Moyer serves as the supervisory logistics management specialist of the Predictive Logistics Section at the Tank-automotive & Armaments Command (TACOM). He previously served as chief of plans within the TACOM G-3 and as an integrated logistics support manager within the Joint Program Office Mine Resistant Ambush Protected vehicles. He was an honor graduate of both the basic and advanced 915A Automotive Maintenance Warrant Officer Courses at the Technical Logistics College. He is pursuing a Bachelor of Science degree in data management and analysis at Western Governors University.
This article was published in the Summer 2023 issue of Army Sustainment.