Situation:
As global supply chains expand, firms are increasingly vulnerable to natural and man-made disasters. With subsidiaries spread across disaster-prone regions, multinational enterprises (MNEs) face significant operational and financial risks. Policymakers and corporations lack robust tools to quantify disaster exposure and resilience over time and across countries.

Task:
The objective was to design a data-driven framework to measure how disasters affect firm performance globally. This required combining geospatial analysis with machine learning to quantify exposure, classify disaster impact, and build firm-level resilience profiles.

Action:

• Data Integration:
  • Compiled data on major disasters since 1990 across 7 categories (biological, climatological, geophysical, hydrological, meteorological, sociopolitical, technological).
  • Geolocated subsidiaries of the world’s largest MNEs and adjusted exposures by ownership proportion.
• Distance Calculation:
  • Calculated proximity between subsidiaries and disasters using the Manhattan Distance Formula (accounting for city-grid-like travel paths instead of straight-line distances).
  • Incorporated disaster-specific characteristics (type, severity, category).
• Machine Learning & Clustering:
  • Applied neural networks to assign weights to disaster types and predict impact radii.
  • Used cluster analysis to group firms by resilience and exposure patterns.
• Risk Profiling:
  • Generated type-specific disaster exposure metrics for each subsidiary.
  • Built enterprise-level risk dashboards linking exposure with operational and sustainability outcomes across time and geography.

Result:

• Developed a novel methodological framework combining geospatial analysis, Manhattan distance metrics, and ML to quantify disaster risk.
• Produced granular resilience profiles for MNEs, enabling comparison across industries and countries.
• Revealed how different disaster types uniquely disrupt firm operations, with implications for global investment and supply chain planning.
• Provided actionable insights for corporate risk management and disaster preparedness strategies worldwide.