Graph-Based Emergency Routing System

Made by Viren Singh

Hub Incident Your location Selected route Estimated (no road data)
Priority City --
Risk Weight --
Stress Mode --
1
Input

Hazard severity, affected people, rescue hubs, incidents, and selected route stress.

2
Process

Build a weighted graph where stress increases travel cost and reduces route capacity.

3
Output

Rank feasible paths by risk-aware cost and show the safest first dispatch route.

System Status Backend ready. Click Run Decision.

Live data sources will appear here after the optimizer runs.

Live Location → Nearest Hospital Click "Use My Live Location" to find the closest hospital hub to where you are right now.

This looks up real nearby hospitals from OpenStreetMap based on your live GPS location, anywhere in the world. It only falls back to this app's own North Bay, California dataset if no OpenStreetMap hospital data is found nearby. This is a routing demo, not an emergency dispatch service - in a real emergency, call your local emergency number first.

Weighted graph routing Ready

Route Ranking

0 nodes - 0 edges

Active Incidents

0 incidents
Incident City People Severity Risk weight

Evaluation: Baseline vs ResQNet

--
Metric Without system With ResQNet

Scaling + Stress Test

Not run
Research Model

Graph-Based Emergency Routing System with Risk-Aware Path Optimization

Graph model Hubs and incidents are nodes. A directed edge exists when a hub can reach an incident inside the regional route limit.
Risk-weighted routing Each edge cost combines travel time, severity, demand load, route stress, and hazard type. Higher stress directly increases route cost.
Capacity allocation Edmonds-Karp max-flow sends limited rescue capacity through feasible hub-to-incident edges.
Decision output The first route is the lowest-risk feasible allocation after stress and capacity are applied, not just the shortest visual line.