The Use of an Artificial Neural Network to Predict Australia’s Export Air Cargo Demand

Abstract

Abstract:In this paper an Artificial Neural Network (ANN) is proposed for predicting Australia's annual export air cargo demand. The modelling in the study was based on annual data from 1993 to 2016. The ANN model was developed using the input parameters of world real merchandise exports, world population growth, world jet fuel prices, world air cargo yields (proxy for air cargo costs), outbound flights from Australia, and Australian/ United States dollar exchange rate and two dummy variables, which controlled for the strong cyclical fluctuations in air cargo demand which occurred in 2003 and 2015. The artificial neural network (ANN) used multi-layer perceptron (MLP) architecture that compromised a multi-layer feed-forward network and the sigmoid and linear functions were used as activation functions with the feed forward-back propagation algorithm. The ANN was applied during training, testing and validation and had 8 inputs, 1 neuron in the hidden layer and 1 neuron in the output layer. The data was randomly divided into three data sets; training, testing and model validation. The best-fit model was selected according to four goodness-of-fit measures: mean absolute error (MAE), mean square error (MSE), root mean square errors (RMSE), and mean absolute percentage errors (MAPE). The highest R-value obtained from the ANN model is 0.97844. The results suggest that the ANN model is an efficient tool for predicting Australia's annual export air cargo demand.

Identifying the key influencing factors for the growth of air cargo demand

“…The final list of factors for the detailed survey and analytical hierarchy process (AHP) analysis is as follows: outbound flights (Glenn Baxter, 2018); load factor (Iman Mohammadian, 2019); number of low-cost carriers (Iman Mohammadian, 2019); airport competition (Shuojiang Xua, 2019; Daniel and Suh, 2019); avg. ticket price (Daniel and Suh, 2019); connecting passenger share (Daniel and Suh, 2019); per capita income (GDP) (Daniel and Suh, 2019); exchange rate (Glenn Baxter, 2018); national price level (Wadud, 2013); industry growth (Totamane et al , 2014); monthly inflation rate (Totamane et al , 2014); world airfreight yield (Glenn Baxter, 2018); world jet fuel price (Glenn Baxter, 2018; Iman Mohammadian, 2019; Wadud, 2013); cargo demand (Shuojiang Xua, 2019; Wadud, 2013; Totamane et al , 2014); cargo demand growth rate (Totamane et al , 2014); world merchandise exports (Glenn Baxter, 2018); pandemic; legal limitations; annual enplaned export air cargo tonnage (Glenn Baxter, 2018); population domestic (Daniel and Suh, 2019; Iman Mohammadian, 2019); world population (Glenn Baxter, 2018); world population growth (Glenn Baxter, 2018); passenger demand (Feng Jin, 2020; Daniel and Suh, 2019; Shuojiang Xua, 2019); passenger turnover (Shuojiang Xua, 2019); domestic total turnover (Shuojiang Xua, 2019); international total turnover (Shuojiang Xua, 2019); Hirschman–Herfindahl index (Iman Mohammadian, 2019; Daniel and Suh, 2019); no. of connected airports (Chiara Morlotti, 2021; Paolo Malighetti and Scotti, 2019); no.…”

“…Typically, air transportation plays a significant role in social and economic development. In the airline industry, air passenger and cargo demand forecasting values are a vital element for the airport, airline and airfreight managers to make timely operation and decision-making plans (Feng Jin, 2020; Glenn Baxter, 2018). Hence, the main objective of this paper is to identify the key influencing factors for the growth of air cargo demand and understand the factors needed to focus on when developing models to identify the air cargo demand.…”