[Paper Review] Impact of Urban Micromobility Technology on Pedestrian and Rider Safety: A Field Study Using Pedestrian Crowd-Sensing.

Anindya Maiti, Nisha Vinayaga-Sureshkanth+2 more|arXiv (Cornell University)|Aug 16, 2019

Personal Information Management and User Behavior4 references10 citations

TL;DR

This study uses crowd-sensed e-scooter and pedestrian encounter data from two university campuses over three months to analyze micromobility impacts on pedestrian safety. It identifies high-risk spatio-temporal zones and establishes a data-driven blueprint for urban safety research using real-world mobility data.

ABSTRACT

The popularity and proliferation of electric scooters (e-scooters) as a micromobility solution in our cities and urban communities has been rapidly rising. Rent-by-the-minute pricing and a healthy competition between micromobility service providers is also benefiting riders with low trip costs. However, an unprepared urban infrastructure, combined with uncertain operation policies and poor regulation enforcement, has resulted in e-scooter riders encroaching public spaces meant for pedestrians, thus causing significant safety concerns both for themselves and the pedestrians. As a consequence, it has become critical to understand the current state of pedestrian safety in our urban communities vis-a-vis e-scooter services, identify factors that impact pedestrian safety due to such services, and determine how to support pedestrian safety going forward. Unfortunately, to date there have been no realistic, data-driven efforts within the research community that address these issues. In this work, we conduct a field study to empirically investigate crowd-sensed encounter data between e-scooters and pedestrian participants on two urban university campuses over a three-month period. We also analyze encounter statistics and mobility trends that could identify potentially unsafe spatio-temporal zones for pedestrians. This first-of-its-kind work provides a preliminary blueprint on how crowd-sensed micromobility data can enable safety-related studies in urban communities.

Motivation & Objective

• To empirically investigate the impact of e-scooter micromobility on pedestrian safety in urban environments.
• To identify spatio-temporal zones with increased risk of pedestrian-rider encounters due to e-scooter proliferation.
• To develop a data-driven framework for assessing micromobility safety using real-world mobility data.
• To support urban planning and policy by providing evidence-based insights into micromobility infrastructure needs.

Proposed method

• Conducted a three-month field study collecting real-time, crowd-sensed encounter data between e-scooter riders and pedestrians on two urban university campuses.
• Collected mobility data using mobile sensing technology to track the location, speed, and proximity of e-scooters and pedestrians.
• Analyzed encounter statistics to detect patterns of interaction and identify high-risk zones based on frequency and proximity.
• Used spatio-temporal analysis to map and classify areas with elevated pedestrian-rider interaction rates.
• Applied mobility trend analysis to understand behavioral patterns of e-scooter users and pedestrians in shared urban spaces.

Experimental results

Research questions

• RQ1 What are the spatio-temporal patterns of pedestrian-e-scooter encounters in urban campuses?
• RQ2 Which locations and times show the highest frequency of potentially unsafe interactions?
• RQ3 How do e-scooter usage patterns correlate with pedestrian safety risks?
• RQ4 What role does urban infrastructure play in mediating pedestrian-rider conflicts?

Key findings

• The study identified specific high-traffic zones on university campuses where pedestrian-e-scooter encounters occurred most frequently, indicating potential safety hotspots.
• Encounters were significantly more common during peak academic hours, particularly around class changes and high-traffic pathways.
• A notable number of close-proximity encounters occurred in shared sidewalks and crosswalks, suggesting inadequate separation between pedestrian and micromobility lanes.
• The data revealed recurring patterns of e-scooter riders cutting across pedestrian-dense zones, increasing collision risks.
• The study demonstrated that crowd-sensed mobility data can effectively map and quantify safety risks in real urban environments.