|
...flows of terrorism but focus on two defensive measures--deployment of physical resources, an example of an averting action, and issuance of warnings, an example of amelioration. The purpose of this paper is to help create a better understanding about the purpose and effectiveness of terror warnings and resource deployments. We believe that this is the first paper to model joint decisions about warnings and resource deployments to minimize the costs of terrorism. By developing a theoretical model of the interplay between resource deployment, terror warnings, and the information available to decision makers we create a structured way to think about these defensive mechanisms and identify when and how they can be used most effectively to provide for public security.
Physically deploying security forces in the field can have several benefits. If the deployment is very visible it can serve as a sign that the government is "doing something" to protect the public and is a visible sign that perhaps the government is anticipating an attack. Similarly, these deployments may deter terrorists from striking. Clearly, the physical presence of security forces can also prevent or limit the extent of a terrorist attack. Physical resources are also expensive, and their cost effectiveness has much to do with the intelligence available on likely threats. Placing a battalion of the New York State National Guard around the Statue of Liberty will not be very effective if terrorists are planning an attack on Los Angeles International Airport. Likewise, if the U.S. government believes that there is an imminent attack planned on a fast food restaurant in the continental United States, deploying a squad of armed troops at every McDonalds and Burger King in the country will be a very expensive exercise. Physical deployments may also reveal defensive procedures if terrorists are probing defenses with feints.
Since March 2002, the United States has adopted a system of color-coded terror warning levels. At any point in time, the Department of Homeland Security (DHS) states the current assessment of the likelihood of a terrorist attack on the United States. These assessments are based on information and analyses from the various intelligence agencies of the U.S. government. In Israel, the security services regularly report on the number of active terror threats they are tracking (see, for example, Dudkevitch 2004). In Israel it is also not uncommon to hear news reports that refer to recent security operations as related to some previous warnings (Myre 2004). Warnings can serve many of the same purposes as physical deployments. Public warnings can create the perception that government authorities are active and informed in a struggle that is often conducted in a secretive way. Public warnings can also deter terror attacks. If a terrorist cell, in the midst of planning an attack, comes to believe (as a result of the warning) their plans have been partially revealed, they may abort them. Warnings can also serve as a mechanism to directly aid in the prevention of terrorist attacks and the reduction in the damage caused by terrorist attacks. The first goal is achieved by increasing the alertness of security personnel at possible targets. The second goal is clearly expressed by the DHS in its description of its alert system. In its literature, the DHS describes the steps that all government agencies should take to prepare their facilities and employees at different levels of threat, with similar guidelines for the general public. Raising preparedness means that if important infrastructure is damaged, government and civil society can continue to function thereby minimizing the damage of the attack. It can also mean canceling or altering activities so that fewer people are present at likely targets, again minimizing the damage from an attack.
Compared to physical deployments, warnings have an important cost advantage. Given the modern communications technologies available, disseminating a warning across a broad geographic area is relatively inexpensive. However, the effectiveness of warnings is limited by their credibility. Pate-Cornell (1986) and Pate-Cornell and Benito-Claudio (1984) model the optimization of warning systems. In particular, they provide a structure for understanding how warning systems work and can be evaluated. Pate-Cornell (1986, p. 223) characterize the costs and benefits of a warning system as being a function of "(1) the quality of the signals issued and (2) the response of the people to whom warnings are directed." These two factors determine the trade-off between failures to issue alarms when something bad happens and the issuance of false alarms. If the public does not believe the warning it will not take the precautions dictated by the raised alert. This holds true for security personnel as well whose work is typically quite monotonous. Therefore, while warnings may be relatively cost effective they cannot be used indiscriminately or their credibility and utility will be diminished.
In this paper, we analyze two types of warnings--private and public. Private warnings are defined as warnings issued to security forces to boost their alertness and preparedness for possible attacks. Public warnings are defined as warnings issued to the general public to similarly boost their alertness and preparedness for possible attacks. Being public, these warnings also serve as warnings to security forces and are observed by terrorists, and as such, may affect terrorist behavior. The economic analysis of terrorism, and responses to it, has found that substitution is an important phenomenon in terrorist behavior (Enders and Sandler 1993). Defensive actions that make one type of target more difficult to attack lead terrorists to different targets. Terrorists are also forced to make intertemporal substitutions shifting attacks from one time period to another in response to attacks on them by security forces. Enders and Sandler (2004) give an overview of the development of this literature. Therefore, in this paper, when we model public warnings we include the possibility that terrorists will shift their attacks to periods in which there is no public warning.
In [section]2, we develop a conceptual model of terrorism risks and defenses using influence diagrams to create a broad structure for the main issues facing a decision maker. In [section][section]3 and 4, we develop and analyze mathematical models that capture the key elements of the conceptual model described in [section]2. While the mathematical models are further abstractions of the situation faced by society and the decision maker, their formalism enables us to investigate the issues more deeply and precisely than the conceptual model. Our goal is to improve our understanding of the interaction between warnings and physical deployments and to identify prescriptions for decision makers. Given the abstraction of the models, we should not expect decision makers to adopt the prescriptions without modification. However, we should expect the prescriptions to shape and spur discussion and further investigation of these important issues. In [section]3 we formulate a model of private warnings and physical defenses. We analyze this model to develop insight into how governments can best utilize warnings and physical defensive resources. We also compare the performance of our model to an alternative heuristic demonstrating the importance of integrating warning and physical deployment decisions and effective intelligence management. In [section]4, we extend our model to include public warnings and develop rules of thumb for the optimal use of public warnings. In particular, we show how expectations about terrorist responses to public warnings can influence the optimal use of public warnings. We conclude in [section]5.
2. Conceptual Model
To better structure our thinking about terrorism defenses, we now develop conceptual models of long-term and short-term risk analysis. These conceptual models will form the basis for our mathematical modeling and analysis in the following sections. We begin by considering a single potential target and then expand our discussion to a spectrum of multiple targets.
When analyzing the long-term risk of terrorism to a particular target, several factors are important: the probability of an attack, the difficulty of doing substantial damage to the target (its hardness), and the economic and political value of the target. In Figure la, we use an influence diagram to depict how these factors work together to determine the cost of terrorism from the perspective of a single target.
[FIGURE 1A OMITTED]
The total cost of terrorism is comprised of damage from attacks and the cost of defensive measures taken. Defensive measures are in addition to the inherent hardness of the target. For example, a shopping mall, designed for open access to the general public, is not as difficult to damage as a nuclear power plant that has restricted access and many safeguards to make it robust. A shopping mall can be hardened by installing concrete barriers that prevent vehicle bombs from getting too close or by installing metal detectors in its entrance ways. Such measures combined with the inherent hardness of the target will influence the damage an attack causes. The damage caused by an attack is also influenced by the value of the target. Damage caused to a high-value target is by definition more costly to society than damage caused to a low-value target. The value of a target is determined by its economic significance as well as its political or symbolic significance. Loss of life has both economic and political significance. Finally, attack damage is influenced by the type of attack, where "no attack" is a possible type of attack. The type of attack is of course uncertain but is itself influenced by the hardness of the target and its value because we assume that the terrorists themselves perform a cost-benefit analysis when selecting a target. Terrorists want to cause significant damage but want a high probability of success.
The analysis outlined above, when applied to all potential targets in a geographic area, will yield what we can term a risk profile for that space. Several insurance companies are currently developing such risk profiles to determine insurance coverage criterion and parameters (see Risk Management Solutions 2005). The purpose of the risk profile is to assess the expected damage costs of terrorism to each target over the lifetime of the insurance policy.
A short-term perspective assumes that the outcome of the long-term risk analysis outlined above has been conducted and operationalized and that its outcome is known by the decision maker and to a large extent the terrorist groups as well. In Figure 1b, we use an influence diagram to depict how intelligence information influences decisions about short-term defensive measures.
Given the long-term risk profile, there is a current state of the world that influences the short-term likelihood of attack and type of attack. Intelligence data gives the decision maker imperfect knowledge of this state that influences his decision making. For example, a long-term risk profile, as developed by insurance companies, may indicate that the set of high-rise office buildings in Manhattan are high-risk targets for a truck bomb. However, security surveillance tapes may indicate suspicious activity by small groups of men filming specific buildings in the financial district. Such information merged with the long-term risk profile would yield an updated probability of an attack on high-rise office buildings in the financial district. Looking at Figure 1b, this means that the probability of an attack on a specific target is influenced by the long-term risk for that target and the current intelligence data. Target value and...
NOTE: All illustrations and photos
have been removed from this article.
More articles from Management Science
How much did the Liberty shipbuilders forget?, June 01, 2007
Category management and coordination in retail assortment planning in ..., June 01, 2007
Analysis and comparison of queues with different levels of delay infor..., June 01, 2007
Service performance analysis and improvement for a ticket queue with b..., June 01, 2007
Performance bounds for flexible systems requiring setups., June 01, 2007
Looking for additional articles?
Search our database of over 3 million articles.
Looking for more in-depth information on this industry?
Search our complete database of Industry & Market reports by text, subject, publication
name or publication date.
About Goliath
Whether you're looking for sales prospects, competitive information, company
analysis or best practices in managing your organization,
Goliath can help you meet your business needs.
Our extensive business information databases empower business
professionals with both the breadth and depth of credible,
authoritative information they need to support their business
goals. Whether it be strategic planning, sales prospecting,
company research or defining management best practices -
Goliath is your leading source for accurate information.
|
