Know Thy Enemy

[Guest]

Know the enemy: mathematical models: the latest weapons against urban insurgencies.


War planners for decades have used computer simulations to prepare for future battles.

The Iraq experience, however, convinced commanders that they need new and improved ways to cope with the complex social environments and insurgent behaviors that were poorly understood before U.S. forces invaded Iraq.

"The Defense Department is asking for models of social agendas, of social behaviors," said Rob Goff, manager of defense operations at Alion Science and Technology Corp. The company develops simulations and models for military war games.

"A lot of work is going on in that area," said Goff.

Simulations that require accurate representations of foreign cultures are tough to do, he said. The challenge is even harder when the models have to predict the interaction between different cultures in harsh conditions, Goff noted.

Military leaders, for example, may want to know what happens when two cultures are thrown into a poverty stricken city--how they would relate to each other based on religion, economics, ethnicity and other factors.

"The hard part is defining the social behaviors among cultures," Goff said. To make simulations believable, they have to truly reflect the characteristics and nuances of specific ethnic and religious groups.

A recent example of the difficulties of modeling social behaviors is a pandemic simulation that Alion created for the Office of Naval Research. "We didn't think it would be that hard, but it is turning out to be quite a chore" because of the complexities of human behavior," Goff said. "Lots of things go into modeling the effects of a pandemic episode other than people catching the bird flu.

"When you try to lay all that out, and make it come true in a simulation, the multiplying effects are just so hard to predict ... We know people may die. But what will happen to the water supply, the food supply, the availability of hospital beds? It's a tremendously tough database to build."

Simulations that can help military commanders anticipate how foreign cultures react to specific actions are in high demand, said George Stone, Alion's senior scientist. "The Defense Department is setting aside funding for these technologies," he said. "A lot of the work is pretty new and everyone is just trying to understand it."

Many experts view these projects as yet another example of the Defense Department over-relying on technology to solve every problem, but military officials argue that these tools are valuable aids and not a substitute for human judgment.

"We want to help commanders evaluate courses of action, and the reactions and perceptions of the population," said Navy Rear Adm. Dan W. Davenport, director of joint experimentation at U.S. Joint Forces Command in Suffolk, Va.

A team of analysts from the command recently was sent to Afghanistan to collect data, not only for future models and simulations, but also for current operations.

"Having the right expertise and the right information is absolutely essential," Davenport said in an interview. In addition to simulation experts, the Joint Forces Command team included anthropologists and sociologists, he said.

Models and simulations, no matter how accurate, will only go so far, Davenport cautioned. "They are just to help inform the commander. They can't make decisions for us."

The simulations that are being used by U.S. commanders in Afghanistan are helping them better understand the potential consequences of their actions, said Dave Ozolek, executive director of Joint Forces Command's experimentation directorate.

Unlike traditional simulation technologies that focus on predicting who will win the war, these newer models are far more sophisticated because they can fairly quickly deliver "what-if" scenarios and help commanders anticipate outcomes based not just on military tactics but also on political, economic and social circumstances.

In Afghanistan, commanders who previously may have been skeptical of experimental technologies came to value how these simulations helped them make decisions, Ozolek said. "In this case, it can put a new capability into the hands of a commander who is desperate for a new operational approach to support his mission."

[ILLUSTRATION OMITTED]

The technology that underpins these simulators is a far cry from the Lanchester equations that have dominated military war gaming for the last 30 years.

Frederick Lanchester, a British mathematician, created formulas to demonstrate the power relationships between opposing forces. These equations were fairly easy to apply to conventional warfare but are not relevant to irregular military operations such as counterinsurgencies.

A more suitable technology for simulating current conflicts is a computer model developed by Purdue University, known as the "synthetic environment for analysis and simulations." SEAS allows researchers to try out their models or techniques in a realistically detailed environment. The simulations gobble up breaking news, census data, economic indicators, and climactic events in the real world, along with proprietary information such as military intelligence. Iraq and Afghanistan computer models are the most highly developed and complex. Each has about five million individual nodes that represent entities such as hospitals, mosques, pipelines and people.

Joint Forces Command analysts like SEAS because of its ability to simulate the non-kinetic aspects of combat, such as the diplomatic, economic, political, infrastructure and social issues, Ozolek said.

The SEAS technology originated at Purdue University a decade ago. It was funded by the Defense Department, Fortune 500 companies, the National Science Foundation, the Century Fund of the state of Indiana, and the Office of Naval Research. Originally, it was developed to help Fortune 500 companies with strategic planning. It was also used to model the population of the United States that is eligible for military service to help the Army recruit potential soldiers.

In Iraq last year, the SEAS technology served as the basis of the Urban Resolve 2015 war game, said Ozolek. "We were really exploring how we change the conditions of the urban environment so the urban insurgencies don't exist."

The latest war gaming efforts in Iraq and Afghanistan mark the first time that Joint Forces Command engineers have taken the experimentation into a combat zone. In the past, that would have been frowned upon, Ozolek said. "We've always been concerned about the risk of taking experimental ideas into real-world operations."

But Army Gen. Dan McNeill, the top commander of U.S. troops in Afghanistan, said he wanted the experiments to continue beyond the initial plans, because he found the models useful for planning operations, according to Ozolek.

A team of 10 engineers from Joint Forces Command works in Afghanistan but most of the modeling computations are done in Suffolk, where 30 analysts crunch data. "Our analysts can look at things that are impossible to look at in a forward area because they are so complex," Ozolek said.

These war planning models do not fall into any of the traditional categories of simulations that typically are used by the military--virtual, constructive or live. This is more about analysis of real-world data, Ozolek said. "We are seeing the first flashes of some really important developments in how we do experimentation. It has opened up a whole new field in modeling and simulation for operational support."

In the past, simulations were viewed in isolation as opposed to being part of the operations, he said. Now, "the data we are using in experimentations is the same data that is being used in operations."

The recent war games in Iraq and Afghanistan also helped Joint Forces Command make a case to its overseers at the Pentagon that the money spent on simulations can deliver tangible results.

Five years ago, the command suffered a public-relations blow when it was disclosed that its largest-ever war game--called Millennium Challenge--cost $250 million and several high-level participants at the time questioned the return on the investment.

Urban Resolve, by contrast, was an example of how simulations can assist real-world operations for significantly less money, Ozolek said. In Millennium Challenge, he said, "we had rudimentary modeling and simulation technologies so we relied almost entirely on field, live simulations of the problem ... That cost $250 million, took three years to arrange, tied up 14,000 troops for six months."

Urban Resolve was planned in less than a year, cost $18 million, and required 1,400 people--mostly civilians.

"We still have work to do in operational metrics," Ozolek said. "But this takes us to the next level of sophistication."

Among other problems that continue to plague military war games is the disparity of data standards, he noted. "It's very difficult to access all the information we need. This is a true command-and-control problem, not just a modeling and simulation problem."


http://www.thefreelibrary.com/Know+the+ene......-a0172132705

http://stinet.dtic.mil/cgi-bin/GetTRDoc?AD...oc=GetTRDoc.pdf