Course of Action Forecasting

From MilcordWiki

Overview

Course of Action (COA) - available both as a Web or mobile application - integrates temporal and spatial data from multiple intelligence sources, identifies patterns in historical data, and predicts future actions at specified spatial locations.

Need

In the current operating environment, it is necessary to:

• Identify the adversary in an asymmetric environment
• Understand local demographics and dynamics
• Anticipate adversary tactics
• Anticipate mission risk
• Reduce information overload in command center

Approach

The COA Knowledge Management (CKM) framework extracts entities (people, places) and spatiotemporal patterns (events, relationships) from unstructured intelligence data, categorizes the extracted patterns using a taxonomy model, discovers and manages COA facts using belief nets, and enables the intelligence analyst visualize and explore the extracted facts.

The approach involves:

• Fusing multi-INT data sources
• Learning adversary behavior models
• Using models to forecast risk
• Providing a platform for battlespace intelligence
• Providing tactical intelligence applications such as for route planning

Benefits

• Faster adaptation to adversary tactics due to system learning
• Reduced information overload due to predictive reasoning
• Reduced mission risk due to integration of intelligence data, adversary behavior patterns, and events

Applications

• Military: U.S. Army, Marine Corps
• Civilian: FBI, Intelligence Fusion Centers
• Commercial: Protection of critical infrastructure industries such as utilities, energy, chemical industries
• Competitive Advantages:

• Unlike incident reporting systems, COA learns models and patterns and predicts behavior
• COA brings a higher level of automation to IPB process planning and implementation

References

• Caglayan, A., Burke, D., Eaton, G. and Meessen, Y. (2008) Course of Action Forecasting, Technical Report, DTIC AD Number: ADB338206. US Army TEC, Alexandria, VA.