e-scooter-redesign

GOAL -
Diagnose and mitigate safety-related design considerations within popular e-scooters

E-scooters represent an increasingly popular component within a larger ecosystem of more sustainable, affordable, and portable modes of transportation. Wide-scale, fast adoption of these new modes of transit, however, have presented various concerns.

As scooters occupy a space that has not traditionally been accounted for (it is neither a car nor a bike), understanding of their rules and regulations currently remains ambiguous. Mounting anecdotes related to falls and/or crashes have called into question safety considerations of these vehicles.

This research aims to study major safety issues related to the control issues while operating Spin scooters, diagnose major pain points, and suggest possible solutions to ensure safer streets for all. Emphasis has been made on the task of riding itself, with a task analysis and user studies guiding the direction of the redesign solution(s).

TASK ANALYSIS

I began my direct research by performing a task analysis on the act of riding a Spin scooter from ‘point A’ to ‘point B’, noting any ambiguous areas or near-slips that I experienced. In order to log this information in a more formalized manner, I elected to employ a decomposition method of task analysis; as the overall transportation task could roughly be divided into the overall sequential tasks of (1) acquiring a scooter to ride, (2) riding a scooter to the destination, and (3) completing the riding session/parking the scooter, each of which consisted of various linear sub-tasks.

From my own experience riding the Spin scooters - including going through the process of creating a task analysis - the process of attempting to use turn signals seemed to be the most dangerous aspect of riding. Not only is this how I fell myself, but it turned out to be the same catalyst for participant P-1’s recent injury, which she recounted after our user study. This also seemed to be the area ripe for a design intervention, as reliance on the use of bike hand signals rested on prior knowledge of the conventions of another transportation modality (biking).

With this task analysis sequence working as a framework, I proceeded to use this as a basis to conduct user studies on the act of riding Spin scooters around Pittsburgh, ensuring that observations between participants were standardized.

USER STUDY

‍OVERVIEW
‍Participants were recruited through my academic network; both are fellow MHCI students, and both have become personal friends. In conducting the two think-aloud sessions, recording was infeasible due to the nature of the task. I essentially asked participants to locate a bike and take a ride with me upon our meeting.Both sessions took place on the same day; weather conditions were moderate (about 50 F) yet windy, posing a slight issue regarding dialogue (we had to ride close to one another, both for observation as well as to maintain conversation). Sessions took place in the late morning with ample sunlight, and thus comments related to riding at night were retrospective. Both participants rode with me (individually) for about twenty minutes, thinking aloud as they went. Once our think-aloud sessions ended, I immediately recounted the findings/insights by recording voice memos in my phone. Participants were present during this recording to ensure that information was accurate, and to provide their own commentary as an immediate retrospective analysis.

DEMOGRAPHICS & INITIAL COMMENTS
‍Participant P-1 was a female, 25-year-old student, with extensive experience riding bikes through urban areas. Having grown up in Brooklyn, NY, she frequently biked around New York with family and friends. This experience translated itself well to riding e-scooters, as she demonstrated a high level of situational awareness and familiarity with traffic laws. I often found myself asking her to explain what she seemed to be doing intuitively - including maintaining proper distances from surrounding vehicles and ensuring that she remained on the right-side of the road - demonstrating a clear skill-based automatic behavior.

Participant P-2 was a male, 24-year-old student, who described himself as an occasional bike rider. He had attended high school and college (undergraduate) in Michigan, and had already been familiar with e-scooters from his time living in Ann Arbor, MI. This participant was less familiar with traffic regulations, and thus demonstrated more task errors during ‘Goal 2: Riding Scooter to Destination’ than P-1. The difference in prior, tangential knowledge was one particularly salient factor between the two participants; I inferred from this that past biking experience correlated to decreases in e-scooter riding errors due to existing knowledge of traffic laws/conventions. This may be viewed as a byproduct of the principle of associations; by having spent so much time using hand-signals through bike-riding, repeated associations were formed and thus ingrained into long-term memory, allowing retrieval of this association while using e-scooters.

Participant P-1 User Study
* Please note: participant used scooter to go home after our user study session, and hence I did not observe Goal 3: Completing Ride/ Parking Scooter with participant P-1.

Participant P-2 User Study

REFLECTIONS ON USER STUDIES
Though participants P-1 and P-2 had different backgrounds - and varying levels of familiarity with traffic conventions - both users indicated similar motivations and pain points related to using the Spin scooters. Primary motivations largely had to do with time savings - “I mostly use scooters when I’m running late to places, like when I’m running late to class or to restaurant reservations, for example.” (Participant P-1)
‍
As such, scooter usage was more likely to coincide with stressful circumstances, suggesting overarousal. A plethora of negative performance changes have been correlated to overarousal - particularly in its propensity to engender cognitive tunneling and working memory loss, both of which would further suggest a need to provide more knowledge in the world to hedge against reliance on working memory when engaging in potentially hazardous scenarios (such as attempting to use hand-turn signals while making a turn on the Spin scooters).

SECONDARY RESEARCH

Researchers in Germany published a study in 2020 looking at hand signal usage while riding e-scooters, with particular emphasis on novice riders. Although their controlled experiment - away from real traffic - did find that new riders were more likely to feel safer using hand turn signals after a short learning period, their report still suggested major issues with relying on increased training alone. Many participants expressed that“the e-scooter becomes significantly slower when the hand is taken off the throttle, which makes it more difficult to control the e-scooter when turning right.” (Löcken et al., 2020, pp. 134) Those that did use hand signals “often made rapid gestures that may be easy to miss in a real traffic situation.” (Löcken et al., 2020, pp.134) Though increased training did positively correlate to “perceived safety”, in real-world situations, most novice riders are generally reaching for e-scooters as a fast and fun way to get around their (generally urban)environment. (Löcken et al., 2020, pp. 137)

A more recent article published in Journal of Automobile Engineering confirmed that “[t]he ride of these e-scooters is often perceived to be difficult to control compared to a standard 26-in or 700C wheel safety bicycle.” (Paudel & Fah Yap, 2021, pp. 2552) The paper began by highlighting the current “lack of proper guidelines for small wheel e-scooters,” which they viewed as a particularly relevant issue when viewed in the context of the increasing rate of e-scooter accidents. (Paudel & Fah Yap, 2021, pp. 2553) Devoting much of their research to a mechanical analysis of safety hazards, they did find that “the current designs of most e-scooters are less stable, difficult to control when encountering obstacles, and more sensitive to braking than the safety bicycle,” corroborating the concerns of participants P-1 and P-2 regarding steering controls. (Paudel & Fah Yap, 2021, pp. 2560)

Löcken, A., Brunner, P., & Kates, R. (2020). Impact of Hand Signals on Safety: Two Controlled Studies With Novice E-Scooter Riders. 12th International Conference on Automotive User Interfaces andInteractive Vehicular Applications. Association for Computing Machinery, New York, NY, USA, 132–140. https://doi.org/10.1145/3409120.3410641

Paudel, M., & Fah Yap, F. (2021). Front steering design guidelines formulation for e-scooters considering the influence of sitting and standing riders on self-stability and safety performance. Proceedings of theInstitution of Mechanical Engineers, Part D: Journal of Automobile Engineering, 235(9), 2551–2567.https://doi.org/10.1177/0954407021992176

REDESIGN

At the end of each user study, I asked participants the following question: “If you could choose one design change that would make your experience using the Spin scooters either more enjoyable or safer, what would you propose?”

‍The responses were startlingly similar - and mirrored my own initial intuition.

“It would be helpful if the scooter had a way that I could click a turn signal and I wouldn’t have to lose my balance.” (P-1)
‍
“I think a simple turn signal ... if I use my hands, then I can’t really hold on to the handlebar. Also, I don’t really know what hand signals to use.” (P-2)

The proposed redesign of the Spin scooters thus focuses on adding a series of light indicators as turn signals. These signals would be controlled by a switch below the brake on the left side of the handlebar; mirroring controls commonly found in automobiles, switching the control up would indicate a right turn, while switching the control down would indicate a left turn. Due to the svelte nature of the scooters, I wanted to ensure that these indicators would be visible from any angle (oncoming traffic from the front, rear pedestrians/vehicles, and traffic coming from the right or left sides). In comparison to hand-signals - which are hard to perceive at night - use of indicator lights would also create contrast against the night sky and thus may target issues of conspicuity when riding at night.

Turn Indicator Light Control Switch: The design of this control aims to mimic that found in automobiles, embodying the memory concept of consistency, which facilitates transfer of skill from one scenario to another. Repeated associations of similar controls allow for greater recall when using a related device. When activated, indicator lights would flash on the side of the handlebar correlated to the turn, in addition to linear lights on the sides and back lights.

TRADE-OFFS
Possible trade-offs to this re-design include the potential for accidental activation, as a user’s thumb would likely be resting close to the control switch, should the hand be positioned to readily control the breaking lever. Lights may also be prone to damage, which would necessitate the use of hand-signals when they aren’t working; however, as users become used to this automation, skill loss in terms of acquiring/remembering appropriate hand-signals may gradually be lost completely.

CONCLUDING REMARKS
The issue of Spin scooter safety - while intriguing in theory at first - became of personal interest to me after my own scooter accident. I was disheartened - though not entirely surprised - to learn that friends and fellow classmates of mine had experienced similar issues. By designing-out the particularly hazardous task
of using hand-signals while turning, I feel that my redesign accomplishes the necessity of indicating turning intentions while providing a safer way to control this via a familiar physical control switch. A safer alternative in terms of control, indicating turns via digital light signals also creates a safer environment for all traffic participants, increasing the conspicuity of scooter riders to surrounding vehicles.