Predicting insurgent attacks with a mathematical model

Dec 17, 2009

When bombs and bullets left 37 dead during Friday prayers at a mosque in Pakistan, earlier this month, the insurgency was using the element of surprise. Unpredictability is the hallmark of modern insurgent attacks such as this one. However, the likelihood of such events, their timing and strength can now be estimated and managed before occurring, according to a new study by researchers at the University of Miami. The study entitled "Common Ecology Quantifies Human Insurgency" is featured as the cover of the December 17, 2009 issue of the scientific journal Nature.

The University of Miami (UM) researchers and their collaborators analyzed the size and timing of 54,679 violent events reported in Afghanistan, Colombia, Indonesia, Iraq, Israel, Northern Ireland, Peru, Senegal and Sierra Leone. The findings show that there is a generic way in which humans carry out insurgency and terrorism when faced by a large powerful state force, and this is irrespective of background history, motivation, ideology, politics and location, explains Neil Johnson, principal investigator of the study and professor of Physics at the UM College of Arts and Sciences.

"We have found a unified model of modern insurgent wars that shows a fundamental pattern in the apparent chaos of wars," says Johnson. "In practical terms, our analysis can be used to create and explore scenarios, make predictions and assess risks, for present and future wars."

The study finds a common statistical distribution for insurgency attacks that is significantly different to the distribution of attacks in traditional wars. This finding supports the belief that insurgent wars represent "fourth generation warfare" with different dynamics from conventional wars.

"Despite the many different discussions of various wars, different historical features, tribes, geography and cause, we find that the way humans fight modern (present and probably future) wars is the same," he says. "Just like traffic patterns in Tokyo, London and Miami are pretty much the same."

The unified model of human insurgency establishes a quantitative connection between insurgency, global terrorism and ecology, and this model demonstrates that insurgency behaves like a "soup-of-groups, with no permanent network or leaders, but with common decision-making processes," says Johnson. This "mathematical law of " challenges traditional ideas of insurgency based on rigid hierarchies and networks, explains Johnson.

Interestingly, this model of human insurgency bears a "striking similarity" to models of crowd behavior of financial markets. Therefore, in addition to its academic and practical contribution to understanding and managing human conflict, "it also suggests a possible link between collective human dynamics in violent and non-violent settings," says Johnson.

In the future, Johnson and his collaborators plan to explore what would happen if a third population is added to the analysis, such as a peacekeeping mission; how should they be deployed in order to minimize casualties? They also hope to apply this understanding of conflict to other areas. For example, in the medical field, a virus might be represented by an insurgency, while the state army is the immune system and a blocking drug may be the peacekeepers. "We are starting to explore its applications in other 'wars', such as the personal battle of a patient with cancer," Johnson says.

Explore further: Everglades trail surveyed for cultural artifacts

add to favorites email to friend print save as pdf

Related Stories

The Mathematical Structure of Terrorism

May 22, 2006

The complex patterns of the natural world often turn out to be governed by relatively simple mathematical relationships. A seashell grows at a rate proportional to its size, resulting in a delicate spiral. The gossamer network ...

Tribal war drove human evolution of aggression

Sep 09, 2008

Wars are costly in terms of lives and resources – so why have we fought them throughout human history? In modern times, states may fight wars for a number of complex reasons. But in the past, most tribal wars were fought ...

Climate change could boost incidence of civil war in Africa

Nov 23, 2009

Climate change could increase the likelihood of civil war in sub-Saharan Africa by over 50 percent within the next two decades, according to a new study led by a team of researchers at University of California, Berkeley, ...

Recommended for you

Secrets of dinosaur ecology found in fragile amber

17 hours ago

Ryan McKellar's research sounds like it was plucked from Jurassic Park: he studies pieces of amber found buried with dinosaur skeletons. But rather than re-creating dinosaurs, McKellar uses the tiny pieces ...

User comments : 2

Adjust slider to filter visible comments by rank

Display comments: newest first

brant
not rated yet Dec 17, 2009
And then the computer program decided that you were an insurgent.
Computer modeling is partly responsible for the financial situation that our country is in right now.

And ClimateGate.
Sean_W
1 / 5 (1) Dec 17, 2009
And then the computer program decided that you were an insurgent.
Computer modeling is partly responsible for the financial situation that our country is in right now.

And ClimateGate.


I agree with ClimateGate being a good example of models and simulations being misused. But there are lots of areas where they are valuable in finding hidden properties or explaining known ones. The distinction is whether or not your model is just an AV prop to illustrate your previously held beliefs (as in climatology) or not.

At first I thought this was going to be about geographic profiling which has been used to search for serial criminals. That technology still requires an experienced human expert and can't just be copied on a disk and sent out to all police departments but has still been able to assist law enforcement.

What's learned from this work may actually help gauge security responses but it won't actually predict a specific terrorist event's time or place.