Analyzing Congestion Dynamics in Amsterdam's Shared Spaces: A Multi-Layered Agent-Based Model of Pedestrian and Cyclist Interactions

By Marta Schwarz

Abstract
The concept of shared space has received increasing attention in the realm of urban planning for its potential to enhance safety, improve traffic flow, and create more inclusive, pedestrian-friendly environments by minimizing traditional traffic controls and encouraging social negotiation among road users. This study investigates congestion dynamics in Amsterdam's shared urban space through a multi-layered model incorporating Social Force Theory. Focusing on pedestrian and cyclist interactions, the study aims to simulate traffic patterns across varying densities, for both current and future scenarios. The model integrates conflict management, pathfinding, and behavioral layers to capture the complex, localized interactions characteristic of shared spaces.This is done by employing Agent-Based Modelling (ABM), aiming to find emergent flow patterns in shared spaces. The validation process demonstrates the model's internal consistency, with particular attention to speed and conflict outputs. Findings reveal that increased density elevates conflict frequency and spatial hotspots, providing spatial information on where targeted urban planning interventions should be developed. Additionally, the model offers valuable insights into future mobility scenarios, emphasizing the importance of adaptive design strategies to balance efficiency and safety. This research highlights the potential for multi-layered agent-based modeling to advance the understanding of shared space dynamics and support sustainable urban development.

Keywords: Agent-Based Modelling; Social Force Theory; Shared Space; Microscopic Modelling