A simple mathematical framework is almost certainly out of reach, but evidence-based crime prevention and law enforcement in the past two decades may show us a path for aggregating small to large unit data into a family of conditional models that can be used to drive macro-scale results. I suspect some similar effort is already underway in Iraq and probably has been for sometime; there's certainly a hint whenever newsies report some battalion commander's assessment of challenges and success in his area of responsibility. Thing is I've never seen anything in the public domain about if and how the underlying data filters up the chain of command and is processed.
Bookmarks