International Journal of Scientific Research and Engineering Development

Urban disaster response has a serious problem. Human dispatchers are too slow, agencies don't talk to each other, and resources get wasted when every second counts. This paper introduces EMERGIPLEX — an agentic AI system that automates the entire disaster response chain. It is built as a set of loosely coupled microservices, all coordinated by a central Orchestrator that polls a shared SQLite database for new tasks. Three core agents drive the system: CrowdIntelAgent validates incoming incidents in real time, Resource Alloc Agent uses the Haversine formula to match and assign the closest available unit, and RouteOptimAgent computes the fastest travel path using A* on OpenStreetMap data. Together, they turn a raw incident report into a confirmed, dispatched logistics plan — automatically. A pilot simulation on a real map of Mandya, India, showed a 77% drop in un-serviced time and a 46% reduction in average ETA compared to a traditional Greedy baseline. EMERGIPLEX is a working, scalable blueprint for next-generation emergency response.

Lahari Raj, M S Hithaishini, Mahadevaprasad MG, Sandeep Kumar, Prof. V Chetan Kumar,"EMERGIPLEX: A Resilient, Coordinated Multi-Agent System for Real-Time Urban Disaster Response and Logistics" International Journal of Scientific Research and Engineering Development, V9(3): Page(377-379) May-June 2026. ISSN: 2581-7175. www.ijsred.com. Published by Scientific and Academic Research Publishing.