ETH Mobility Initiative Project MI-01-19 Empirical use and Impact analysis of MaaS: Ergebnisse

Abstract

The Empirical use and Impact analysis of MaaS (EIM) Project focused on the question whether and how Mobility as a Service (MaaS) changes travel behavior (e.g., mode choice) and how we can represent mobility data better to create machine learning applications for the prediction and interpretation of individual mobility. The core of the project was a user study designed around the introduction of yumuv, a new MaaS product launched in Zurich, Switzerland, in Q3/2020. We collected a multi-source dataset (survey data, GPS tracking data, and context data) of 498 users (71 in treatment group, 427 in control group) over a period of 3-month during the roll-out of the MaaS product yumuv. A mode choice analysis showed that the MaaS bundle mainly increased the usage of e-scooters with a small positive effect on public transport usage. Substitutes are mainly trips with personal bicycles as well as personal e-bikes. The bundle had no influence on the use of personal cars. Furthermore, we present a graph-based representation for individual mobility as well as an application to use it for the identification of dataset independent user groups.