Artificial intelligence enhances air mobility planning

The intricate world of air mobility planning, a critical component of national security, is undergoing a significant transformation through the integration of artificial intelligence. The U.S. Air Force’s Air Mobility Command, specifically its 618th Air Operations Center (AOC), is leveraging AI to streamline complex operations, ensuring rapid and efficient global responses.

The 618th AOC, recognized as the Department of Defense’s largest air operations center, manages a fleet of a thousand aircraft. Each day, hundreds of chat messages circulate between pilots, crew, and controllers, coordinating myriad variables from flight routes and fueling times to cargo loading and crew assignments. This vital coordination, traditionally handled through phone and email, has transitioned to chat, opening new avenues for AI-driven enhancements.

“It takes a lot of work to get a missile defense system across the world, for example, and this coordination used to be done through phone and email. Now, we are using chat, which creates opportunities for artificial intelligence to enhance our workflows,” states Colonel Joseph Monaco, Director of Strategy at the 618th AOC.

To capitalize on these opportunities, the 618th AOC is sponsoring Lincoln Laboratory in the development of sophisticated AI tools under a project named Conversational AI Technology for Transition (CAITT). This initiative is part of a broader Air Force modernization effort, the Next Generation Information Technology for Mobility Readiness Enhancement (NITMRE).

At the core of CAITT is Natural Language Processing (NLP), an AI discipline that enables models to comprehend and process human language. Courtland VanDam, a researcher in Lincoln Laboratory’s AI Technology and Systems Group, which leads the project, explains, “We are utilizing NLP to map major trends in chat conversations, retrieve and cite specific information, and identify and contextualize critical decision points.” CAITT encompasses a suite of tools designed to leverage NLP for operational efficiency.

Among the most advanced tools is topic summarization. This feature automatically extracts trending subjects from chat messages, presenting them in a user-friendly display that highlights critical conversations and emerging issues. For instance, an alert might read: “Crew members missing Congo visas, potential for delay.” The system provides the number of related chats, summarizes key discussion points in bullet form, and links directly back to the specific chat exchanges. Colonel Monaco emphasizes the tool’s importance, noting, “Our missions are very time-dependent, so we have to synthesize a lot of information quickly. This feature can really cue us as to where our efforts should be focused.”

Another crucial tool in production is semantic search, a significant upgrade to the existing chat service’s search engine. Unlike the current system which requires exact keyword matches, the semantic search tool allows users to pose questions in natural language, such as inquiring about the cause of a specific aircraft delay, and receive intelligent, relevant results. “It incorporates a search model based on neural networks that can understand the user intent of the query and go beyond term matching,” VanDam adds.

Further tools are currently under development to enhance mission planning. These include functionalities to automatically add relevant users to chat conversations based on their expertise, predict ground time needed for specific cargo unloading, and summarize key processes from regulatory documents to guide operators in mission development.

The CAITT project originated from the DAF–MIT AI Accelerator, a collaborative effort involving MIT, Lincoln Laboratory, and the Department of the Air Force (DAF) to advance AI algorithms and systems. Lieutenant Colonel Tim Heaton notes, “Through our involvement in the AI Accelerator via the NITMRE project, we realized we could do something innovative with all of the unstructured chat information in the 618th AOC.”

As Lincoln Laboratory researchers continue to refine CAITT prototypes, they have initiated the transition process to the 402nd Software Engineering Group, a software provider for the Department of Defense. This crucial step will integrate the advanced AI tools into the operational software environment utilized by the 618th AOC, marking a new era of AI-enhanced air mobility planning.