Comparing Urban Elements Effects on Objective and Subjective Cycling Safety

Abstract

Understanding factors influencing cycling safety is critical for promoting sustainable mobility. Cyclists require both objective safety (protection from crashes) and subjective safety (feeling safe). Previous research has examined these two safety components separately, i.e., they typically focus on one safety type at a time, or it has analyzed the effects of a limited number of urban elements on either type of safety. This study integrates street-level imagery and spatial data to characterize and compare how urban cycling contexts influence both objective and subjective safety. Using machine learning (XGBoost + SHAP), we analyze 120 urban elements' impacts on 3,850 cycling crash locations in Berlin, Germany. Results show that some elements have similar effects on both safety types (e.g.: buildings and amenities), while others affect only one type (e.g.: crosswalks and truck presence), and some have opposing effects (e.g.: roadside features). These findings highlight the importance of aligning urban planning with both actual and perceived safety needs, as interventions targeting one type may inadvertently harm the other.