PedNStream release: scalable pedestrian network flow simulator

PedNStream is an open-source, Python-native simulator for macroscopic pedestrian network loading, submitted to arXiv on 1 Jul 2026 by Weiming Mai, Dorine Duives and Serge Hoogendoorn. The framework implements the Link Transmission Model (LTM) and adds stochastic link dynamics and a utility-based route choice to support closed-loop control testing.

What is PedNStream and how does it work?

PedNStream is a modular macroscopic pedestrian network simulator built on the Link Transmission Model, implemented in Python and released as open-source. It extends LTM-based pedestrian models by incorporating stochastic link dynamics that capture diffusion and activity-induced variability, and replaces dynamic user equilibrium route choice with a utility-based formulation suited to uncertain, intervention-driven settings.

The framework includes built-in controller interfaces for interventions such as gating, flow separation, and route guidance. That modular design lets controllers interact with macroscopic link states and route choice modules during simulation runs, enabling closed-loop experiments where interventions change demand or link behavior.

How did the authors validate PedNStream and measure scalability?

Validation used a staged evaluation: synthetic scenarios, real-network experiments, a closed-loop control case study, and a runtime analysis. Synthetic scenarios verified mechanisms including queue formation, spillback, congestion dissipation, and adaptive rerouting. Real-network experiments assessed large-scale behavior and consistency with observed pedestrian counts.

The paper presents a runtime analysis that quantifies scalability. The arXiv submission lists the manuscript as 13 pages with 14 figures (arXiv:2607.01021), and the evaluations are structured to demonstrate both behavioral realism in counts and the practical performance needed for network-scale studies. The closed-loop case study demonstrates controller integration with the simulation framework.

Why it matters

Pedestrian management at city and venue scale needs simulators that run efficiently and support controller testing; PedNStream targets that gap by combining macroscopic LTM performance with stochastic dynamics and controller hooks. That combination enables experiments where interventions like gating or route guidance can be evaluated in a closed-loop setting rather than only in open-loop or microscopic tools.

By being Python-native and open-source, PedNStream also aims to lower the barrier for practitioners and researchers who need repeatable, scriptable testbeds for large-scale pedestrian traffic management and control algorithm development.

What to watch

Look for the code release and documentation linked from the arXiv entry for arXiv:2607.01021 and for follow-up studies using the built-in controller interfaces for gating, flow separation, and route guidance. Upcoming validations that report numerical runtime benchmarks and broader real-network deployments will be the clearest signals of adoption.