Urban Logic
[01]
Transport Simulation
[Desciption]
An agent-based simulation evaluating the cross-district spillover effects of traffic flows on high street performance and accessibility. Uses AI-enhanced behavioural modelling to assess how traffic decisions in one district propagate through adjacent urban corridors.
[Model Objective]
This model simulates Saudi morning peak traffic (5:30–8:00 AM) across a 3×3 grid of urban districts, using Riyadh as its reference context. It quantifies how school service deficits generate excess arterial traffic — when a district lacks a school type, households must cross main roads to reach adjacent districts, compounding congestion beyond what baseline through-traffic alone produces. While grounded in Riyadh's urban structure, the underlying logic applies to any city where uneven service distribution drives cross-district travel demand.
Research Prototype
2026
Cross-District Traffic Impact Model
Know more
The model monitors peak arterial car counts, average trip times by type (local, cross-district, and baseline through-traffic), and the real-time congestion state of all roads. Once calibrated to a specific urban context, it supports sensitivity analysis — revealing how incrementally adding or removing a school type shifts traffic load, changes peak timing, and affects overall network performance.
[Model Components]
3×3 district grid with Saudi-style dual-lane divided arterials · Household agents with school-age children across 6 gender-segregated school types (Elementary / Middle / Secondary × Boys / Girls) · Multi-stop chained trip optimisation · Baseline through-traffic with 4 arrival distribution modes · 16 signalised intersections with configurable cycle times · Real-time congestion feedback loop · Stacked area chart dashboard
[Model Results]
The model monitors peak arterial car counts, average trip times by type (local, cross-district, and baseline through-traffic), and the real-time congestion state of all roads. Once calibrated to a specific urban context, it supports sensitivity analysis — revealing how incrementally adding or removing a school type shifts traffic load, changes peak timing, and affects overall network performance.
[Model Components]
Agent-Based Modelling · Dijkstra Shortest-Path Routing · Greenshields Speed-Flow Congestion Model · Truncated Normal Departure Time Distribution · Seeded PRNG (reproducibility) · A/B Scenario Comparison
[Tags]
["ABM", "AI", "Transport", "Urban Mobility"]