Future mobility demand estimation based on sociodemographic information: A data-driven approach using machine learning algorithms

Abstract

Estimations of the future mobility demand are highly valuable for policymakers, transportation planners, and the automotive industry. Knowing mobility patterns allows for targeted and optimized decarbonization of the transport sector. This work provides a model order reduction approach for clustering mobility demand according to characteristic population groups that share similar travel behavior. Using Swiss household travel survey data and machine learning algorithms, the methodology developed in this paper allows for extrapolating future mobility demand based on socio-demographic information.