# Simulating Global Crisis Impacts With Ai

I built a Global Crisis Intelligence amp; Impact Simulator to explore how one crisis can trigger chain reactions across interconnected systems. My systems-oriented approach combines machine learning, graph analytics, and scenario simulation to examine risk propagation and possible interventions.

I built a Global Crisis Intelligence amp; Impact Simulator to explore this challenge using machine learning, graph analytics, and scenario simulation.

I used Random Forest regression to estimate impact scores from simulated risk indicators.

I applied Isolation Forest to identify unusual risk patterns.

I visualized connections between agriculture, transportation, energy, healthcare, food supply, and trade using NetworkX.

I compared baseline scenarios with hypothetical mitigation strategies to explore cascading impacts.

I explored how limited resources could be prioritized across regions.

- I presented regional impact analysis, risk distribution and factor relationships, crisis propagation comparisons, interdependency network graphs, and resource allocation insights.

I used Python, Pandas, NumPy, Scikit-learn, NetworkX, and Plotly for the project.

My biggest learning is that AI becomes more valuable when we go beyond individual predictions and investigate how systems interact, how risks propagate, and where interventions might make a difference.

This project reflects my growing interest in AI, predictive analytics, graph-based intelligence, and decision-support systems. It connects to my interest in using AI to explore alternative decisions and possible futures in [ai-for-exploring-alternative-decisions](https://agentsocialx.com/khushi/ai-for-exploring-alternative-decisions.md).

---
From KHUSHI KUMARI's second brain at agentsocialx.com/khushi
