Observation results on pedestrian-vehicle interactions at non-signalized intersections towards simulation

•We propose an evidence-based approach for developing an agent-based simulation tool.•The approach is exemplified in a pedestrian crossing scenario in which a video-recorded observation was performed.•Results focus on traffic volumes, Level of Service, drivers’ compliance, crossing decisions.•Crossing behaviour is composed of three phases: approaching, appraising and crossing.•Analyses highlighted and quantified significant differences among adult and elderly pedestrians. The modelling and simulation of pedestrian-vehicle interactions is an open challenge for both research and practical computational solutions aiming at supporting urban/traffic decision makers and managers. The social cost of pedestrian-car accidents pushes the development of more effective and expressive computational models and simulation systems, integrating analytical knowledge, data and experience about pedestrian-vehicle interactions. The paper presents the results of a significant data gathering campaign focused on the considered phenomenon and aimed at supporting the extension and calibration of an existing model. A video-recorded naturalistic observation was executed in an area of the City of Milan (Italy) characterized by a significant presence of elderly inhabitants and by a high number of road accidents involving pedestrians in the past years. The achieved results include both macroscopic and microscopic indicators about pedestrian-vehicle interactions at the observed non-signalized intersection, with reference to: (i) traffic volumes; (ii) Level of Service; (iii) drivers compliance to pedestrians right of way on zebra crossing; (iv) age-driven pedestrian crossing behaviour, focusing on motor skills (speeds and trajectories) and decision making (accepted time gap to cross). Since elderlies represent one of the most vulnerable pedestrians in terms of road accidents, data analysis is focused on comparing results among two samples composed of adults and aged pedestrians. Results showed that crossing behaviour is characterized by three main phases: approaching, appraising (evaluation of the distance and speed of oncoming vehicles) and crossing. Moreover, results showed a significant difference in the crossing behaviour of adult and elderly pedestrians. The final objective of the research is to support the development of a microscopic agent-based tool for simulating pedestrian-vehicle interactions at non-signalized crosswalks.