Modeling and Simulation Adds Insight on Terrorism

December 2001
By Dr. Roger Smith

New technologies can unravel the complexities of global rogue organizations.

The United States has been using simulations for decades to explore the capabilities of its military forces and train soldiers to perform their missions better. In the war against terrorism, however, this technology can come out of the training centers and into the operations centers to support the country and its allies in fighting this new type of war and enhance homeland security.

While today’s databases hold a wealth of information about various aspects of an adversary’s operation, collating that knowledge in a multidimensional format can help decision makers meet today’s challenges and predict unseen but looming dangers. Now, as the adversary has become more complex, leaders need more powerful tools to help them manage, understand and penetrate this complexity. Modeling and simulation adds insight to information.

Dealing with this new threat will require the application of all types of military and intelligence assets, including computer systems for managing, sifting and correlating data. This could include the application of systems dynamics models, combat engagement simulations and analytical war games that manage complexity and discover relationships and meaning in data that is spread across multiple domains and distributed along a causally related timeline.

When used as analytical and decision support tools, models and simulations are like databases that change automatically in response to relationships between old and new information. While a database is a way to organize, store and search for data, a simulation is a dynamic tool for rearranging, combining, changing and exploring new configurations of data from one minute to the next. This makes it a potentially invaluable tool for predicting future terrorist actions based on past and present events and conditions.

The professional simulation community has not heavily focused its resources and ingenuity on a complete analysis of the terrorist threat. As a result, a need exists for models that capture all aspects of terrorism and look for warning signs of future actions. Combat models that allow Special Forces to rehearse attacks against terrorist strongholds do exist. Intelligence models that identify the logical and physical relationships among members of a terrorist cell are available. But, what is lacking is a digital model that includes the military, economic, political, security and legal facets of the threat.

A suite of tools that describes this world is conceivable, and the search for such a tool should become a primary focus of organizations involved in studying, understanding and countering terrorist actions. The White House executive order that established the Office of Homeland Security calls for that office to “prevent terrorist attacks within the United States.” Dynamics models operating on the data the United States has gathered about terrorist groups would help prevent future attacks by identifying the conditions for their emergence and predicting the forms they will take, their most likely locations and the possible timing.

A terrorist organization may consist of one or more command nuclei, multiple field cells around the globe, host nations that allow it to operate, operational assets such as weapons and communications equipment, financial assets, and sympathetic groups and individuals that provide support. All of the activities and interests of a terrorist organization cannot be represented in a single homogeneous model; the problem is too big and too diverse. Such a complex group is better represented in a suite of models that exchange information that is relevant in one dimension of the problem and that also impacts another dimension.

A diverse organization presents many unique features that are intertwined. Each is a different facet of the problem that must be described and captured in a particular way within a simulation. Like the real-world organization, each facet has specific relationships with the others, and these connections are essential to understanding how the entire organization works. For instance, legal actions to tie up financial assets may have a direct effect on the ability of a field operative to purchase airplane tickets and pay for living expenses. Political and economic sanctions against a host nation may hinder its ability to establish and maintain training camps, thus reducing the number of terrorists available for missions. Humanitarian assistance may decrease the level of malaise within a country, shrinking the terrorists’ pool of potential recruits and limiting the types of missions that these organizations can launch.

Simulations are some of the best tools for exploring the dynamic relationships between objects and activities. Traditionally, databases and spreadsheets have been used to capture a static snapshot of data. Add-on products allow these databases to sift, analyze and compare data in an attempt to identify important conditions that should be highlighted. However, this is not as powerful as creating a model specifically designed to mimic and predict terrorist actions. These models tie hundreds or thousands of these database snapshots together over time by identifying the relationships between the objects and actions involved. Algorithms that represent combat intelligence on an adversary’s movement, human behavior, logistics support and other activity can be used to track current events and predict what is possible in the future. The result is a moving picture of the past and present that can be used to peer into the future.

Although the benefits are many, developing the technology may be a challenge. A weakness of all models and simulations has been the lack of an ability to discover or distill relationships between objects and events into specific algorithms. In theory, if these relationships were perfectly known, then the model would perfectly predict future states and events. In fact, some scientists view the world as a giant simulation in which all relationships are known and the “world simulation” uses these to predict the future by creating it. Such an idealistic perspective encourages practitioners to discover these hidden relationships so that digital simulations can approximate the same calculations and provide a window into the future before it arrives.

Many organizations use databases, spreadsheets and other tools to capture and manage information about their mission. The human mind then serves as the modeling engine for exploring the dynamic and future states of that data. This type of simulation leverage is being applied to identifying the best operational patterns for commercial airlines and interstate trucking. The number of objects, variables and interactions in such large systems makes it impossible for a human to analyze and optimize the entire system.

The military also has been a proponent of analytical and training simulations, but has not yet focused on the terrorist threat. Models are specially constructed to capture and animate a subset of the real world; however, researchers are a long way from being able to create a single simulation that can look at all of the dimensions involved in a terrorist organization. Therefore, the best approach is to create multiple simulations, each focusing on one aspect of the organization. These can then be linked together such that the information exchanged from one dimension to another is only that which influences the second dimension.

Numerous military laboratories already contain multiple linked simulations that form a unified virtual world. However, the purpose of creating multiple simulations has been to spread the computational workload across multiple computers, not to explore multiple dimensions of that world. A suite of counterterrorism simulations would use existing simulation interoperability techniques, but each simulation would represent a unique facet of the problem—military, security, economic, political, legal and humanitarian actions and objects.

A simulation of the physical world would be used to place the command nucleus and field cells on the map. These assets would then become the targets for air strikes, Special Forces attacks, intelligence collection and police arrests. Many combat simulations with these capabilities already exist and can be adapted to this new mission.

A simulation of the legal, financial and political domain would capture the state and dynamic changes of host nations, group financial assets, and the legal status of field cell members. In these, the United States and its allies would apply political and economic pressure to the host nation. Changes to this nation’s status would change the ability of the command nucleus to hide and operate within the physical model. This would enable changes in other models to impact the success of operations in the physical model and vice versa. Models of legal actions against field cells would impact the cell’s ability to enter and operate within the target country. This could directly impact the numbers and locations of physical attacks on targets.

Finally, a model of the cultural, psychological and humanitarian environment could be used to represent the will of sympathetic individuals to support terrorist networks. By applying humanitarian aid to displaced people in the host nation, the United States and its allies may change the ability of that nation to continue to harbor the command nucleus. Aid also may influence global sympathizers to stop providing financial support and shelter to terrorists.

These concepts are an initial outline of the role that models, simulations and decision support tools can play in the war on terrorism. Such a diverse and complex problem has not been captured by these types of tools before and will require significant exploration before it can be captured accurately. If it becomes part of the daily fight against terrorism, such a tool can be constantly changing, improving and becoming more accurate.


Dr. Roger Smith is the vice president of technology at BTG Incorporated.