Japan’s All Nippon Airways (ANA) has become the first airline in the world to implement an AI-based real-time turbulence prediction service. The effort stems from a collaboration with Keio University, which in 2023 evolved into the Tokyo-based venture BlueWX. The collaboration marks a significant step forward in enhancing flight safety, passenger comfort, and operational reliability. Work began in 2019, followed by trial runs from 2021, and full operational implementation as of early August 2025.
ANA’s AI system is a result of years of extensive testing and validation., Developed using deep learning techniques and a decade of turbulence data, the turbulence prediction model has an 86% accuracy rate. It was also validated in trials involving 2,500 pilots. The AI is now embedded within ANA’s weather infrastructure and feeds information directly into flight-planning and in-flight displays.
Hiroyuki Kometan, Executive VP of Operations at ANA, commented on how turbulence prediction has posed difficulties for aircraft operations for years. According to him, the new AI-enabled system offers greater predictability and a smoother flight experience.
BlueWX CEO, Kaz Watanabe, emphasised that the launch illustrates the AI system’s precision and points to broader application across other airlines, as turbulence continues to rise due to global warming.
Overcoming the Unpredictability of Turbulence
Turbulence remains a topic of heated discussion among aviation professionals. According to the US NTSB, turbulence contributes to around 30 to 50 percent of aviation incidents. While turbulence is rarely fatal, it does carry the potential of inflicting serious injuries to both passengers and crew members on board a flight. It can also trigger unscheduled inspections, delays, or cancellations.
Recent industry incidents highlight the high stakes involved. For instance, on May 1, 2024, Singapore Airlines’ Boeing 777-300ER hit turbulence over Myanmar, leading to one passenger dying and over 100 passengers being injured. The flight was diverted to Bangkok.
Similarly, on July 30, 2025, Delta DL56, which was en route from Salt Lake City to Amsterdam, encountered severe turbulence over Wyoming. 25 crew and passengers were injured, and the flight made an emergency landing in Minneapolis.
Many such cases can be taken into consideration when studying the impact of turbulence on flight operations. For instance, on August 28, 2024, a Boeing 737 from Cancun to Chicago hit turbulence, leading to seven injuries, and was later diverted to Memphis. Likewise, a Lufthansa aircraft was en route from Buenos Aires to Frankfurt on November 11, 2024, when it was stuck with turbulence mid-Atlantic. Around 11 minor injuries were reported. A Qantas flight faced a similar situation when it was diverted to Brazil en route from Madrid to Montevideo. While 30 received care, 10 were hospitalised.
These incidents illustrate turbulence’s unpredictability and harm potential. A tool offering earlier, more accurate warnings could shift outcomes toward safer, more controlled operations.
Enjoying the article?
Follow us and never miss an update on the aviation industry.
The Future of AI in Aviation
Historically, artificial intelligence has seen a broader usage in advisory tools and optimisation systems, such as fuel efficiency, scheduling, and predictive maintenance. ANA’s new system looks at its integration into flight planning and onboard decision support. AI technologies have mostly stayed on the ground, where they have been used for analyzing maintenance logs, route planning, and predictive engine wear. However, ANA’s interactive AI is designed to be used by pilots, shaping flight choices during active operations.
ANA’s AI-powered turbulence prediction system is a positive step forward in the ongoing transformation of air travel. The development represents the airline’s commitment to using cutting-edge technology to enhance both safety and comfort. BlueWX will continue the development of wind-speed and direction forecasts to guide route optimisation, aiding efforts toward carbon neutrality by 2050. The system will see broader integration as the carrier looks for more responsive, safer, and efficient operations.
As more airlines begin to recognize the value of AI in improving operational efficiency and passenger experiences, the aviation industry is likely to see widespread adoption of similar technologies soon. This move by ANA signals a shift - AI becoming part of the sensory and decision-support layer in real-time flight management.
