The U.S. Air Force improves aircraft readiness with AI and predictive maintenance solutions

Keeping military aircraft and related equipment up and running is critical to the U.S. Air Force’s (USAF) mission — to fly, fight, and win. Across 59 air bases in the US and more than 100 airfields overseas, the USAF is responsible for more than 5,400 aircraft with an average age of 28 years. A delay in operations when hardware is out of commission costs valuable time, resources, and labor. For the USAF, being able to predict when aircraft need to be grounded for repairs or updates helps maintain mission readiness while lowering maintenance costs.

To address this need, the USAF has employed predictive maintenance solutions, powered by Amazon Web Services (AWS). These solutions use artificial intelligence (AI) algorithms and data analytics tools to monitor operations, detect anomalies, and predict possible defects or breakdowns in equipment before they happen. By making effective use of data analytics solutions in the cloud, defense customers can automate data collection and quickly repair equipment before problems occur.

The Air Force’s Rapid Sustainment Office (RSO) plays a critical role in addressing maintenance challenges. RSO’s mission is to optimize warfighter readiness by exploiting technologies to revolutionize sustainment operations. It supports maintenance, repair, overhaul, and supply chain management for the USAF’s extensive fleet of military aircraft. RSO established the Condition Based Maintenance Plus (CBM+) program office to apply AI and machine learning (ML) to optimize fleet maintenance, increase aircraft availability, and minimize downtime.

As the USAF’s legacy maintenance systems were unable to scale to address myriad use cases and needs, RSO and the CBM+ program office turned to AWS Partner C3 AI to develop a new solution. C3 AI and the USAF co-developed the Predictive Analytics and Decision Assistant (PANDA) – a suite of predictive logistics applications that integrates AI and ML across a variety of aircraft maintenance data. As the data and insights generated through PANDA are mission-critical, C3 AI launched the solution in AWS GovCloud (US). AWS GovCloud (US) is designed to host sensitive data and regulated workloads that address the most stringent U.S. government security and compliance standards. By hosting PANDA on AWS GovCloud (US), C3 AI and the USAF are providing the flexibility to architect secure cloud solutions that address DoD compliance requirements, including the DoD Cloud Computing Security Requirements Guide (SRG) for Impact Levels 2, 4, and 5.

“PANDA makes it easier for the Air Force to maintain fleet readiness through AI-based insights that allow the department to not only easily manage the upkeep of their complex aircraft, but stay ahead of critical maintenance needs,” said C3 AI CEO Thomas M. Siebel. “PANDA ingests massive amounts of sensitive, defense-related data daily, so we chose a cloud provider that has tools to safeguard information vital to national security and has a high compute power that can operate in many different environments — and AWS GovCloud (US) checks all of those boxes.”

Today, PANDA is used to deliver insights to a variety of organizations that work with USAF aircraft. Through data analysis, PANDA helps identify aircraft systems and components at risk of failure, provides evidence for the need for proactive maintenance, and helps to ensure availability of the right parts at the right place at the right time. PANDA is built on the C3 AI Platform and C3 AI Readiness, and is comprised of eight mission applications that support more than 775 active users.

Since the USAF deployed PANDA in AWS GovCloud (US), it makes more accurate predictions about needed repairs and replacement parts, reducing over-maintenance and increasing aircraft readiness. Using ML models has accelerated the time it takes to extract, analyze, and act on relevant maintenance data. PANDA is used to actively monitor thousands of components, and has helped the USAF successfully predict hundreds of failures before they happen. Today, the toolkit has scaled to monitor more than 3,000 aircraft across 16 aircraft platforms.

Most importantly, with the use of predictive maintenance solutions provided by C3 AI and AWS, more aircraft are ready for flight and the USAF can focus on what’s most critical – the defense mission.

Learn more about AWS GovCloud (US) and cloud computing for defense.