Share with your friends


Analytics Magazine

Viewpoint: Analyzing Decisions to Counter Terrorism

Spring 2008


by Randolph W. Hall

As the defining issue of his presidency, the decision to invade Iraq is without question the most important event of George W.Bush’s political career.Had he used decision analysis in making this momentous choice, as proposed in “Should the U.S. Have Attacked Iraq” (OR/MS Today, December 2006), perhaps the process would have been more systematic, and perhaps Bush would have found a way to retain public support for his eventual choice. But had he used decision analysis, I hope that Bush would have gone well beyond exploration of the odds that Iraq did or did not possess weapons of mass destruction (WMD),for the success or failure of the mission should certainly not be judged by this question.

WMD is a nebulous term encompassing a range of weapons, some of which are widely available and easy to manufacture, others of which are not particularly destructive compared to conventional weapons. The “weapons” (box cutters and the like) used by the terrorists who perpetrated the 9/11 attacks were so conventional that they could have been purchased at the local grocer, and terrorists in the past have never depended on non-conventional weapons to further their cause. The key question here should not have been whether Iraq possessed WMD, but whether it offered a unique conduit of weapons that would empower terrorists. Were the weapons suspected of being in Iraq unavailable elsewhere? Did Iraq not only possess these weapons, but was it also collaborating with terrorists? Were these weapons so threatening that they substantially added to the danger of catastrophic terrorism?

More important than WMD, the success or failure of the Iraq mission might be judged by a less public argument for the invasion, that of installing a democracy in the center of the richest oil producing region of the world, and a hoped for domino effect influencing the surrounding nations.Here we had a truly visionary goal, but one that was harder to explain to the public than, for instance, the imagery of Secretary Rice’s “mushroom cloud” consequence of Iraq’s alleged WMD.We also had a considerably more difficult goal to accomplish, one where military action would have been one of many possible tactics and which, in retrospect, was not very effective. In this context, the uncertainty preceding the Iraq invasion was not so much whether WMDs were present, but whether military intervention would bring about long-term stability.

I raise these points to frame the question of how decision analysis might be used to improve our response to terrorism, and to illustrate three challenges: adversarial behavior, interdependencies and politics.

Terrorists are by nature agile and adaptive adversaries; they operate clandestinely and succeed by surprise. They cannot be stopped by protecting individual targets, or by preventing access to any one group of weapons because alternatives will always exist.And history does not provide a very satisfying prediction for the likelihood of future terrorist actions.While it might be noble to envision decision trees that specify the likelihoods of alternative futures, these probabilities are extraordinarily difficult to estimate.

Our strategies for fighting terrorism are highly interdependent because our finite resources are insufficient to protect against all threats. By selecting Iraq for military action, we have made that one part of the war on terrorism our first priority for federal budgeting and allocation of military assets.We have also made it our first priority for the attention of the commander-inchief, as well as our congressional leaders. Given our limits, the decision to invade Iraq should not be judged just on whether it achieved its mission; it should also be judged on whether Iraq should have been picked over its alternatives, such as addressing the threats posed by Iran or North Korea, or more directly targeting terrorists themselves.

Lastly, the decision to go to war, as with most strategic choices in terrorism, must be placed in its political context. President Bush could not have authorized the invasion without first gaining support from Congress. Reluctantly, he also sought support from the United Nations, and more deliberately he sought active participation from our allies. The steps to convince others to collectively support a choice are in many ways distinct from the steps of making the choice, making it hard to decipher the true goals from past public statements.

All told, it is no easy task to encapsulate the Iraq invasion within a decision tree.Yet decision analysis could have been used to illuminate the question of how best to counter terrorism; to elicit alternative courses of action, anticipate the response of adversaries and specify the goals. Retrospectively, we might look back at the Iraq war and evaluate the decision within the narrow lens of how accurate were our predictions that Iraq possessed WMD. But the decision was considerably more complex, revolving around two complementary goals, one immediate – reducing the threat posed by al-Qaeda – and one long-term – building stability and democracy in the Middle East. And, perhaps most importantly, were we trying to solve the right problem?

As the O.R. community engages in the big challenges of the day, such as how to best counter terrorism, we should not be tempted to squeeze the questions into the framework of our existing models, because the problems may go beyond what we have seen in the past.We should, however, use our skills to stimulate the type of creative thinking envisioned by the 9/11 Commission, in which we reduce the threat of terrorism through better understanding of the motivations and tactics of our adversaries, as well as the available alternatives for reducing the threat.

Randolph Hall is vice provost for research advancement at the University of Southern California, and was founding principal investigator for CREATE, the Center for Risk and Economic Analysis of Terrorism Events.



Does negative political advertising actually work?

While many potential voters dread campaign season because of pervasive negative political advertising, a new study has found that negative political advertising actually works, but perhaps not in the way that many may assume. The study “A Border Strategy Analysis of Ad Source and Message Tone in Senatorial Campaigns,” which will be published in the June edition of INFORMS’ journal Marketing Science, is co-authored by Yanwen Wang of the University of British Columbia in Vancouver, Michael Lewis of Emory University in Atlanta and David A. Schweidel of Georgetown University in Washington, D.C. Read more →

Meet Summit, world’s most powerful, smartest scientific supercomputer

The U.S. Department of Energy’s Oak Ridge National Laboratory on June 8 unveiled Summit as the world’s most powerful and smartest scientific supercomputer. With a peak performance of 200,000 trillion calculations per second – or 200 petaflops – Summit will be eight times more powerful than ORNL’s previous top-ranked system, Titan. For certain scientific applications, Summit will also be capable of more than three billion billion mixed precision calculations per second, or 3.3 exaops. Read more →

Employee engagement a top concern affecting customer experience

Employee engagement has surfaced as a major concern in delivering improvements in customer experience (CX), with 86 percent of CX executives in a Gartner, Inc. survey ranking it as having an equal or greater impact than other factors such as project management and data skills. “CX is a people issue,” says Olive Huang, research vice president at Gartner. “In some instances, the best technology investments have been derailed by employee factors, such as a lack of training or incentives, low morale or commitment, and poor communication of goals." Read more →



INFORMS Annual Meeting
Nov. 4-7, 2018, Phoenix


Making Data Science Pay
July 30-31, 12:30 p.m.-5 p.m.

Predictive Analytics: Failure to Launch Webinar
Aug. 18, 11 a.m.

Applied AI & Machine Learning | Comprehensive
Sept. 10-13, 17-20 and 24-25

Advancing the Analytics-Driven Organization
Sept. 17-20, 12-5 p.m. LIVE Online

The Analytics Clinic: Ensemble Models: Worth the Gains?
Sept. 20, 11 a.m. -12:30 p.m.


CAP® Exam computer-based testing sites are available in 700 locations worldwide. Take the exam close to home and on your schedule:

For more information, go to