Bikepath Scanner: Improving usability through user evaluation
Authors
Sangroula, Sujata ; Maskey, Pawan ; Basnet, Abishek
Term
4. term
Education
Publication year
2026
Pages
62
Abstract
This thesis examines how usable and safe the BikePath Scanner app is for cyclists. The app detects bike-path hazards using the phone’s camera (on-device object detection) and logs locations with GPS. Rather than asking whether detection works, the study focuses on whether people can operate the interface while riding without distraction. A mixed-method evaluation was used: expert review (heuristic inspection), task-based tests with participants speaking their thoughts (think-aloud), and a standard usability questionnaire, the System Usability Scale (SUS). The study found nine usability and accessibility issues in three areas: clear feedback for split attention, preventing mistakes under cognitive load, and making information readable at a glance. The SUS score was 72.8 ("Good"), showing solid usability but also that users did not always trust the system’s reliability. All issues were addressed through iterative redesign. The project also includes a dashboard for municipalities to review reported hazards. These results contribute to HCI research on mobile use in safety-critical situations and on digital tools for assessing cycling infrastructure.
Denne afhandling undersøger, hvor brugbar og sikker BikePath Scanner-appen er for cyklister. Appen registrerer farer på cykelstier med telefonens kamera (on-device objektgenkendelse) og gemmer positioner med GPS. I stedet for at teste om selve registreringen virker, spørger studiet, om man kan betjene grænsefladen under kørsel uden at blive distraheret. Der bruges en blandet metode: ekspertgennemgang (heuristisk inspektion), opgavebaserede tests hvor deltagerne siger deres tanker højt (think-aloud), samt et standard spørgeskema for brugbarhed, System Usability Scale (SUS). Studiet fandt ni problemer med brugbarhed og tilgængelighed inden for tre områder: tydelig feedback ved delt opmærksomhed, forebyggelse af fejl under kognitiv belastning, og information der kan aflæses med et hurtigt blik. SUS-scoren var 72,8 ("God"), hvilket viser solid brugbarhed, men også at brugerne ikke altid oplevede systemet som pålideligt. Alle problemer blev løst gennem iterativ redesign. Projektet omfatter også et dashboard til kommuner, som kan gennemgå indmeldte farer. Resultaterne bidrager til HCI-forskning om mobil brug i sikkerhedskritiske situationer og digitale værktøjer til vurdering af cykelinfrastruktur.
[This apstract has been rewritten with the help of AI based on the project's original abstract]
Keywords