Cycling speed profiles from GPS data: Insights for conventional and electrified bicycles in Switzerland

Understanding cycling speed dynamics is crucial for effective transportation planning and infrastructure development. This study analyzes GPS-based cycling speed profiles in Zurich, Switzerland, focusing on conventional bicycles, e-bikes (25 km/h), and speed pedelecs (45 km/h). Using GPS data from 351 cyclists, we examine the influence of socio-demographic factors (age, gender, BMI), road infrastructure, gradients, and weather conditions on cycling speeds. Our findings reveal that speed pedelecs achieve the highest speeds, frequently exceeding residential speed limits, raising questions about their classification and integration into urban mobility networks. Machine learning models identify road gradients, BMI, and age as key determinants of cycling speed. Additionally, results show that e-bikes and speed pedelecs experience longer intersection delays. These insights offer valuable contributions to urban transport policies, cycling infrastructure planning, and traffic modeling, ensuring safer and more efficient mobility solutions. • Analysis of speeds for bikes, e-bikes (25 km/h), and s-pedelecs (45 km/h). • GPS data enable detailed analysis of cyclist speed profiles and variability. • Infrastructure, bike type, and individual factors influence cycling speeds. • S-pedelecs reach highest speeds and face longer average delays at intersections. • Machine learning shows gradient, age, and BMI best explain variation in speeds.