Skip to main content
To KTH's start page To KTH's start page

Shared e-scooter usage patterns

Analysis of app and trip data

Time: Tue 2024-03-19 13.00

Location: U51, Brinellvägen 26, Stockholm

Video link:

Language: English

Subject area: Transport Systems Transport Science

Doctoral student: Boel Berg Wincent , Transportplanering

Opponent: Docent Johan Olstam, Linköpings universitet

Supervisor: Professor Erik Jenelius, Transportplanering, Centrum för transportstudier, CTS; Docent Wilco Burghout, Transportplanering

Export to calendar

QC 240226


There is an urgent need to shift the urban mode distribution towards public transportation, micromobility, and shared mobility as a critical step towards achieving sustainable cities. Micromobility, with shared e-scooters as the main driving force, has sprung up as a promising mode for cities since the late 2010s. The development has been rapid and concurrent, with many changes in policy and operations. Given the novelty of the mode, there is an evident need for planners, policymakers, e-scooter companies, and researchers to understand the usage patterns and potentials of shared e-scooters. This thesis studied shared e-scooter usage patterns by analyzing app and trip data. 

Paper I analyses the access and walking distance to shared e-scooters. The Euclidean access distance was calculated from app and trip data. A process was then developed to estimate map-based walking distance from the Euclidean distance by removing and replacing outliers with an approximated value. Finally, a catchment area for shared e-scooters was presented. The result show that the walking distance is short, the majority walking less than 95 m and 2.1 min.

Paper II evaluated shared e-scooters as a last-minute mode, a mode used as a way to mitigate the risk of late arrival. The study was based on the assumption of a preferred arrival time, which was tested where there was an associated risk of arriving late and when the assumed preferred arrival time shifts. Trip characteristics(speeds, distances, ride times, and trip frequency) of last-minute trips were analyzed and identified. The result show a peak in the number of trips ending before the full hour of the morning. These trips are infrequent on the user level and have lower average ride duration and higher average speeds, indicating an larger share of last-minute trips.