Neural network forecasting for airlines: A comparative analysis

This paper provides the first published research paper on the technique of neural network forecasting as applied to the airline industry and finds the most basic neural network structures provided better forecasts than traditional forecasting methods.

Abstract

This paper provides the first published research paper on the technique of neural network forecasting as applied to the airline industry. It compares this new method with the traditional forecasting techniques (moving averages, exponential smoothing, regression, etc.). The data were provided by a major international carrier. All the methods were compared on the basis of a standard error measure — mean absolute percentage error (MAPE). The results of the study are promising. The most basic neural network structures provided better forecasts than traditional forecasting methods.