Predicting estimated time of arrival for commercial flights (2018)

A KLM Royal Dutch Airlines Dreamliner expects to arrive 20 minutes earlier

Arrival times for commercial flights are currently estimated using deterministic models that fail to account for many variables that affect flight time. Ayhan, Costas, and Samet created a model that leverages these traditionally overlooked variables to make more accurate arrival time predictions.

Why it matters

The estimated time of arrival (ETA) tells you when a flight will probably land at an airport.

That information is useful for passengers and those who pick them up, but possibly even more so for airlines: knowing the ETA of a flight well in advance allows an airline to coordinate personnel and equipment at airports, which minimises turnaround times and lowers costs.

Traditional ETA prediction methods typically use a deterministic approach that only takes the flight trajectory and possibly some of the aircraft’s characteristics into account. However, actual time of arrival is also affected by factors like wind, temperature, and congestion, so a prediction that ignores these variables is not likely to be very accurate.

How the study was conducted

The authors aimed to create a system that accurately predicts the ETA of a commercial flightIt’s important to note that usually there’s more than one ETA. Passengers will typically only care about the gate arrival time. In this study, the authors only look at the runway arrival time (i.e., when the wheels touch the ground), since they do not take taxiways and aprons into account. before it even departs.

The system makes use of several types of data:

Multiple models were constructed and ranked based on their predictive performance. Two of the models make use of boosting methods:

The system’s performance was evaluated with 10 major flight routes in Spain, using 11 different machine learning algorithms and one algorithm based on averages of historical flight times for the same route and period.

What discoveries were made

The resulting system achieves a higher accuracy than EUROCONTROL’s ETA prediction system.

Algorithms

The table below shows the root mean squared error (RMSE), averaged over all routes for each algorithm. The boosting methods appear to work best, and more consistently at that.

Method Algorithm Average RMSE Standard deviation
Traditional Historical average 4.454057 0.910959
Linear Linear regression 5.224831 0.840323
Lasso regression 4.204375 0.858625
Elastic net regression 4.153771 0.805265
Non-linear Classification and regression trees 4.660715 0.453537
Support vector regression 3.886390 0.733385
k-nearest neighbours 3.643647 0.751386
Ensemble Adaptive boosting 🥈 3.364734 0.531285
Gradient boosting 🥇 3.346209 0.461617
Random forest regression 3.498223 0.512965
Extra trees regression 🥉 3.491921 0.503401
Recurrent neural network Long short-term memory 4.298340 1.438574

More detailed results are available in the original article.

Features

The table below lists the relative importance of the top 10 features. It’s clear that meteorological data are invaluable if you want to make accurate ETA predictions.

Rank Feature Score
1 Arrival airport 1.0
2 Atmospheric pressure 0.67854
3 Atmospheric wind speed 0.66231
4 Atmospheric wind direction 0.65224
5 Atmospheric humidity 0.63331
6 Atmospheric temperature 0.61314
7 Airport congestion rate 0.53212
8 Sector congestion rate 0.31153
9 Flight number 0.29192
10 Aircraft type 0.13221

The important bits

  1. The highest ETA prediction accuracy can be achieved using gradient boosting and adaptive boosting
  2. Meteorological data and congestion rates are important predictors for the ETA of a commercial flight