Validating Large Scale Simulations of Socio-Political Phenomena

There is a growing recognition that simulations representing social and political phenomena are very useful to today’s military to support planning and execution of operations associated with irregular warfare. While many technical challenges are being addressed by DARPA, social science simulations still present an overarching problem: methods and capabilities to show that a simulation is working correctly for an appropriate class of applications are not sufficiently developed for a sponsor/user to confidently apply it in an operational environment. It is EBR’s belief, supported by related experience, that the credibility of such simulations can be established if 1) the use is appropriate for the type of simulation; 2) the simulation has been designed to be validatable for that use; and 3) a complementary set of validation and verification capabilities and techniques are available. This work will provide DARPA with a vision for validatable simulations, including appropriate use classes, simulation capabilities needed to enable validation, and promising tools and techniques to assess the degree to which a simulation is sufficiently credible to support the identified classes of applications. This research will identify technical challenges associated with this vision and selected a set of solution options for development in Phase II.