Simulicity IA

Simulicity IA is an interactive 3D urban disaster simulator that helps users explore how cities respond to flooding and how different interventions affect evacuation outcomes. Users can generate fictional cities or import real-world neighborhoods, run scenarios, and compare measured impacts on residents, buildings, and roads. The project integrates Meta’s Muse Spark models through the Meta Model API to translate natural-language descriptions into validated city-generation parameters, propose interventions based on simulation metrics, narrate individual citizens’ simulated situations, and produce readable emergency reports. An intervention agent connects AI proposals with simulation feedback, allowing users to explore drainage, early warnings, and building reinforcement. Meta models provide the language and reasoning layer, while a deterministic simulation engine calculates hazards, evacuation, and outcomes. This separation makes AI proposals testable against measured results rather than treating generated text as evidence. Local fallbacks keep the application usable when AI is unavailable. Built towards “Push the Frontier,” Simulicity IA explores how AI can make complex urban systems easier to understand and experiment with. It is an experimental demonstration, not a calibrated forecasting system or certified emergency-planning tool.