ABSTRACT
Recent urban practice has highlighted the need for safe and efficient pedestrian movement in urban settings. Urban planners and policymakers are tasked with ensuring that individuals can navigate cities while avoiding crowding and minimizing safety risks. Among the various proposed solutions to achieve these planning objectives, enhancing public space adaptability to changing circumstances, such as fluctuations in pedestrian demand and crowding, has garnered significant interest. However, constrained municipal budgets limit the feasibility of large-scale, capital-intensive upgrades to pedestrian infrastructure. As a result, there is an urgent need for a short-term, temporary, and cost-effective strategy to redesign pedestrian public space. This paper introduces a novel tactical urban planning approach, combining evidence-based urbanism. Using a campus digital twin system as an urban simulation platform, this study presents a unique evidence-based planning approach to improve pedestrian spacing on sidewalks by dispersing pedestrians without overly interfering with existing infrastructure. As interventions within existing pedestrian public spaces, one-way sidewalk systems and building entrance control are considered. Through the proposed agent-based simulation of the campus digital twin system, the optimization of their application and the demonstration of its effectiveness are achieved. The outcome of the simulations verifies that these measures can enhance pedestrian spacing by dispersing them without causing substantial adverse effects. While these findings stem from experiments conducted on a particular campus, limiting their generalizability, they still hold significance as they verify the potential for practical implementation of tactical urban planning concepts utilizing a digital twin system for urban environments.
Yang, S., Usman, M., Hassan, A., Sohn, G., Faloutsos, P., & Haworth, B. (2025). Digital twin-based evaluation of tactical interventions for high-density pedestrian environments. Geo-spatial Information Science, 1-19.
https://doi.org/10.1080/10095020.2025.2600195