Fujitsu and Microsoft Collaboration on AI Agents Streamlining Japan Airlines Cabin Crew Workflows

Use case typeAI platformUpdated Jun 13, 2026

Fujitsu and Microsoft partnered to run a pilot program with Japan Airlines Co., Ltd. (JAL) for the "JAL-AI Report," an app powered by generative AI designed for cabin crew. The trial demonstrated that this AI solution simplifies the report creation process and lessens the workload compared to current applications. The app leverages generative AI to assist cabin crew in creating reports about onboard incidents by simply entering keywords and selecting a few checkboxes. It works offline, on-device, edge, and on-premises to ensure usability during flights. Fujitsu collaborated with Headwaters Co., Ltd. to develop this app using Microsoft's Phi, a small language model optimized for offline use, fine-tuned on JAL's past reports. The field trial showed a significant reduction in report creation time and correction needs.

Organization
Japan Airlines
Location
Japan
Published
August 2025

Reported outcomes

Strategic outcomes

New product / capabilityDeveloped an AI incident reporting appSpeed & agilitySimplified report creation workflowNew product / capabilityEnabled offline onboard operationCustomer experience & trustEased cabin crew workload

Primary read

Use case focus

Showing 3 of 4

  • 1AI Agents
  • 2Generative AI
  • 3Edge Computing
Cabin crew reported onboard incidents require complicated and time-consuming report creation.
  • Developed the 'JAL-AI Report' app with generative AI agents that assist cabin crew in generating incident reports efficiently by inputting keywords and selections.
  • App designed to run offline and on edge computing, leveraging Microsoft's Phi small language model fine-tuned on JAL-specific past reports.
  • Used Azure cloud infrastructure as a foundation for development and deployment.
  • Collaboration between Fujitsu and Headwaters Co., Ltd.
  • Significantly reduced report creation time and correction needs.
  • Improved workflow efficiency and eased workload for cabin crew.
Architecture

The architecture involves the 'JAL-AI Report' app running on-device and on edge computing, using Microsoft's Phi small language model fine-tuned with JAL's past reports, integrated with Azure cloud infrastructure. Developed in collaboration with Fujitsu and Headwaters Co., Ltd.

Implementation partners2
Sources & evidence1
Groundedness: 4/5

AI-generated summary. Verify important details with the linked sources before relying on this case.

Explore related AI use cases

Was this useful?

Similar cases