Predictive Modeling of Willingness to Fly on Urban Air Mobility Aircraft

It is predicted that urban air mobility (UAM) services for passenger transportation will reach widespread commercial operations within the next ten years. However, there are a limited number of consumer surveys that support this prediction and not much is known as to the applicability of these surveys beyond their target demographic. This present study uses data obtained from a recent public perception survey about UAM in Canada to predict willingness to fly based solely on socio-demographic and economic characteristics and limited information about past travel behaviors and support for other technologies. Three techniques were employed to generate predictions: ensemble linear regression, a Bayesian network, and an artificial neural network (ANN). The ANN was the most accurate of the three techniques at identifying consumer intentions, choosing the correct response 45.1% of the time on a five-point Likert scale with the accuracy rising to 63.7% when the scale was reduced to three possible outcomes.