CASE STUDY: JAPAN AIRLINES (JAL)
“…Japan Airlines has pioneered a system that goes a long way towards not only improving the information that pilots have about turbulence that might be on their flightpath but also ensuring that that information is received quickly in order to offer the maximum opportunity to avoid the turbulence or take mitigating action to ensure passenger safety.”
W
hile all aircraft encounter some degree of turbulence on most flights, moderate turbulence can make movement around the cabin difficult. Severe turbulence can be more dangerous not so much to aircraft safety (they’re built to withstand it) but to passenger safety and comfort. In this light, any solution that can minimize aircraft exposure to turbulence, by helping pilots avoid the condition, will enhance passenger comfort and safety. With Weathernews, Japan Airlines has pioneered a system that goes a long way towards not only improving the information that pilots have about turbulence that might be on their flightpath but also ensuring that that information is received quickly in order to offer the maximum opportunity to avoid the turbulence or take mitigating action to ensure passenger safety. Perhaps the best way to start this article is with a description of the overall solution (figure 1).
component of this system is a machine learning algorithm developed by Weathernews. This algorithm processes the altitude of turbulence reported to the ground and then automatically notifies nearby aircraft in-flight. So, let us now look at those machine learning algorithms used to process turbulence information and send notifications automatically to aircraft (figure 2).
How to process PIREP using Machine Learning Algorithms (AI) Example:
FL120-020 INC TB2O3 INSIGHT 500 AGL
Components of Jointly Development Solution
Figure 2
Figure 1
There are two main components in this new system. The first one is a solution developed by Japan Airlines. This is the first of its kind in Japan and it uses technology that automatically calculates turbulence indicators. The second
The system was put into operation in December 2020. Machine Learning (ML) technology records the turbulence information shared through the aircraft communication system. After recording, the report is displayed on Japan Airlines’ Operation Management’s reporting system. Turbulence exceeding a certain threshold is automatically sent to aircraft flying within a specific range of the turbulence report position. The threshold and the range can be set by the user. Information such as aircraft position and observation time are included in the report. As a result, pilots will have a better understanding of the risk of turbulence that might affect their flight path.
AIRCRAFT IT Operations • MAY-JUNE 2021 • 35