Mode choice, substitution patterns and environmental impacts of shared and personal micro-mobility

Shared micro-mobility services are rapidly expanding yet little is known about travel behaviour. Understanding mode choice, in particular, is quintessential for incorporating micro-mobility into transport simulations in order to enable effective transport planning. We contribute by collecting a large dataset with matching GPS tracks, booking data and survey data for more than 500 travellers, and by estimating a first choice model between eight transport modes, including shared e-scooters, shared e-bikes, personal e-scooters and personal e-bikes. We find that trip distance, precipitation and access distance are fundamental to micro-mobility mode choice. Substitution patterns reveal that personal e-scooters and e-bikes emit less CO₂ than the transport modes they replace, while shared e-scooters and e-bikes emit more CO₂ than the transport modes they replace. Our results enable researchers and planners to test the effectiveness of policy interventions through transport simulations. Service providers can use our findings on access distances to optimize vehicle repositioning.