KLM deploys AI to reduce in-flight food waste

Overview

KLM uses a machine-learning model called TRAYS, developed with Kickstart AI, to predict how many booked passengers will actually board a flight so its Meals On Board System can calculate the exact number of meals needed. The prediction runs from 17 days before departure until 20 minutes before departure and has cut food waste by up to 63% based on the expected number of passengers per flight, saving more than 100,000 kg of meals a year across flights catered from Schiphol.

The challenge

Not all bookings made result in a passenger on board a KLM aircraft -- depending on the class, between 3 and 5% of booked passengers do not show up (on time) for the flight -- making it difficult to calculate the exact number of meals needed without wasting food.

The solution

KLM developed the AI model TRAYS with Kickstart AI, the first model specifically developed for KLM's catering activities, to predict the number of passengers on board based on historical data. The Meals On Board System (MOBS) receives the expected passenger numbers per flight with separate forecasts for Business, Premium Comfort and Economy classes. The prediction runs from 17 days before departure until 20 minutes before the flight departs, covering the entire catering process from purchasing to loading.

Reported business value

A three-month analysis showed 63% less food wasted compared to catering for every booked passenger. The largest improvement was seen on intercontinental KLM flights from Schiphol, where 2.5 fewer meals (1.3 kg) need to be thrown away per flight, amounting to a saving of 111,000 kg in meals annually across all KLM flights catered from Schiphol.