Abstract
Micro-mobility services (e.g., e-bikes, e-scooters) are increasingly popular among urban communities, being a flexible transport option that brings both opportunities and challenges. As a growing mode of transportation, insights gained from micro-mobility usage data are valuable in policy formulation and improving the quality of services. Existing research analyses patterns and features associated with usage distributions in different localities, and focuses on either temporal or spatial aspects. In this paper, we employ a combination of methods that analyse both spatial and temporal characteristics related to e-scooter trips in a more granular level, enabling observations at different time frames and local geographical zones that prior analysis wasn't able to do. The insights obtained from anonymised, restricted data on shared e-scooter rides show the applicability of the employed method on regulated, privacy preserving micro-mobility trip data. Our results showed population density is the topmost important feature, and it associates with e-scooter usage positively. Population owning motor vehicles is negatively associated with shared e-scooter trips, suggesting a reduction in e-scooter usage among motor vehicle owners. Furthermore, we found that the effect of humidity is more important than precipitation in predicting hourly e-scooter trip count. Buffer analysis showed, nearly 29% trips were stopped, and 27% trips were started on the footpath, revealing higher utilisation of footpaths for parking e-scooters in Melbourne.
Introduction
As cities embrace e-scooters to reduce carbon footprints and traffic congestion, they face a growing "clutter" problem. E-scooters are often left haphazardly, obstructing pedestrians and creating safety hazards. Existing research often takes a "one-size-fits-all" approach, looking at either spatial factors (where people go) or temporal factors (when they go) in isolation, or relying on aggregated data that masks local nuances. To create better urban policies—like where to build parking hubs or how to redistribute fleets—planners need a granular, city-specific understanding of how e-scooters interact with existing infrastructure and weather patterns.
Methodology and Datasets
The researchers employed a multi-pronged analytical approach using data from Melbourne (August to October 2022). The study utilized anonymized e-scooter trip data from the Lime service (~175,000 starts and ~180,000 stops), demographic data from the Australian Bureau of Statistics, land use data from the City of Melbourne, and hourly weather metrics.
- Spatial Regression: Four machine learning models—Negative Binomial Regression (NBR), Support Vector Regression (SVR), Neural Networks (NN), and Random Forest Regression (RFR)—were compared to predict trip density across 350 Statistical Areas. RFR emerged as the most accurate.
- Temporal Analysis: Hourly trip counts were modeled against weather and time-of-day variables.
- GIS Buffer Analysis: 60m buffers were created around Points of Interest (POIs) and 10m buffers around different path types (footpaths, cycle lanes, shared paths) to calculate the percentage of trips interacting with specific infrastructure.
- Multi-Resolution Modeling: Separate models were built for different time windows (e.g., weekday mornings vs. weekend nights) to capture shifting usage patterns.
Key Results and Predictors
The Random Forest model outperformed others in spatial predictions. Analysis of feature importance revealed that population density is the primary driver of high usage. Conversely, there is a strong negative correlation between car ownership and e-scooter usage, suggesting e-scooters serve as a primary mode of transport for those without vehicles rather than just a "last-mile" solution for car owners.
Demographically, the 30-39 age group was the most significant predictor in Melbourne. Temporally, the study found that humidity is a more significant deterrent to riding than actual precipitation. Commuter flow analysis confirmed the e-scooter's role in the morning commute, with a significant trend of trips ending near offices between 6:00 AM and 10:00 AM.
Infrastructure Impact and Footpath Encroachment
A critical finding of the study is the extent of footpath encroachment. The buffer analysis confirmed that footpaths are the primary "parking" infrastructure for shared e-scooters. Approximately 29% of trips ended on a footpath, and 27% started there. This usage significantly exceeds that of dedicated cycle lanes or shared paths, highlighting the need for designated parking solutions to maintain pedestrian accessibility.
Limitations and Future Work
The study was limited by privacy-preserving data rotation, which prevented tracking individual trip trajectories. Additionally, the focus was on a single provider (Lime) over a three-month period. Future research should expand to include multiple providers and longer durations to account for seasonal variations. These models can eventually be used to evaluate the effectiveness of designated parking hubs in reducing footpath clutter.
Summary updated