Factors influencing shared micromobility services: An analysis of e-scooters and bikeshare

Abstract

This study explores how factors, including weather, day of the week, holidays, and special events, influence the trip frequency of two micromobility modes, shared e-scooters and bikeshare, in Louisville, Kentucky. Negative binomial generalized additive models (NBGAM) were estimated to model the trip frequency of each mode. NBGAM provides a rigorous modeling approach that accounts for temporal autocorrelation among variables. While results showed some differences exist between how various factors impact shared e-scooters and bikeshare trips, several similarities emerged between modes. Rain reduced trips for both, reducing bikeshare by 17% and shared e-scooters by 16%. Mondays, Thursdays, Friday, and Saturdays had increased use of both micromobility services though Tuesdays and Wednesdays only saw significant increases in bikeshare ridership. This study contributes to the existing literature in the micromobility realm by quantifying and comparing time-dependent relationships for e-scooters and bikeshare. Results of this study inform how providers distribute vehicles and how cities manage e-scooter policies.