”In collaboration with Kickstart AI. Thanks to the improved forecasts, we have succeeded in reducing food waste by 63 percent. This also reduces the weight on board and therefore fuel consumption. Investing in new technology that contributes to more sustainable development feels good and important,” says Maria Hagelberg, responsible for sustainability issues for Air France and KLM in Northern Europe.
KLM is working with AI to calculate the number of meals needed on board. The reason for this is that not all booked passengers turn up. Depending on the booking class, this can vary between 3 – 5 per cent.
The AI model (TRAYS) is the first to be specifically developed for KLM’s meal handling and it predicts the number of passengers on board based on historical data. The Meals On Board System (MOBS) receives the expected number of passengers per flight based on Business, Premium Comfort and Economy. The forecast starts 17 days before departure and continues until 20 minutes before the flight departs. This means that the most accurate number of passengers is part of the entire catering process from procurement to on-board delivery, and a surplus of meals on board is avoided. The AI model TRAYS was launched at the end of last year by Kickstart AI. The initiative has gathered knowledge from leading companies such as Ahold Delhaize, bol, NS and ING, as well as KLM to develop this model.
A three-month analysis shows that 63 percent less food has been wasted compared to catering calculated for each booked passenger. The biggest difference is noticeable on KLM's long-haul flights from Schiphol, where 2.5 fewer meals (1.3 kg) are thrown away per flight. Over a year, this means that 111,000 kg of meals are saved on KLM flights departing from Schiphol.
