On September 11th, 2001, Rick Rescorla, Vice President for Security at Morgan-Stanley/ Dean-Witter, demonstrated the value of good response planning.  While his identification of the risk associated with the World Trade Center was startling in its accuracy, his ability to translate that knowledge into action was credited for saving for saving the lives of 2,700 of his fellow Morgan-Stanley/ Dean-Witter employees.  As highly populated civilian locations increasingly become the preferred targets for terrorist attacks, the need for information-based threat assessment and response planning increases, as well.  

 “Forewarned, forearmed; to be prepared is half the victory,” Miguel de Cervantes

The same tools that are being used for business intelligence on a daily basis can be used to support safety and security.  Predictive analytics have proven to be an extremely effective approach in the identification, characterization, modeling and prediction of potential threats.  Ultimately, these tools enable law enforcement and security professionals the ability to anticipate, prevent, deter and respond in a more effective manner.

The ability to successfully translate predictive analytics and their promise to the applied public safety and security setting has required domain-specific changes in data preprocessing, including the development of operationally-relevant recoding and variable selection, as well as an emphasis on “operationally-actionable” output.  Specifically, the analytical process needed to be able to accommodate the unique limitations and constraints encountered frequently in the applied public safety and security setting, while generating output that can be immediately interpreted, translated and used by security personnel:  Actionable Mining and Predictive Analysis.  The innovative use of graphics and other novel visualization techniques to translate the results of the analytical process directly to the applied setting has supported the development of output that allows the end user to incorporate their domain expertise, intuition and tacit knowledge in the interpretation of the results; adding significant value to the outcome. 

The benefits of using predictive analytics in the applied public safety and security setting are twofold.  First, the early identification and characterization of a potential threat presents more options for prevention and deterrence.  Targeted prevention strategies also offer a greater return on the public safety investment by supporting the effective allocation of resources.  Personnel, in particular represent an extremely valuable resource in public safety and security.  The ability to proactively place personnel resources when and where they are likely to be needed increases their potential efficacy, while minimizing redundancy and waste.  Moreover, prevention is almost always less expensive than response and recovery, particularly when measured in human terms. 

Second, the use of predictive analytics in the applied public safety and security setting supports information-based response planning.  As highlighted by the Morgan-Stanley/ Dean-Witter experience above, foreknowledge of a potential threat may not be sufficient to prevent it, but the number of lives saved by targeted response planning can be significant. 

The threats we face currently have changed in both nature and potential targets.  In many ways, the frontlines for the war on terrorism have transcended the traditional battlespace and now reside in resorts, transportation systems, critical infrastructure, shopping malls, the financial sector, schools and other places where innocent people congregate.  The solution for information-based prevention, deterrence and response strategies, however, may come from one place – predictive analytics.

McCue has also a book coming out soon that covers this area, as well.  If you want to buy this book click this link.

About the Author
Colleen McCue received a Ph.D. in experimental psychology in 1989. The program of study included significant instruction in advanced and multivariate statistical analysis. Currently McCue works for RTI International. She was till the summer of 2004 the Program Manager of the Richmond, Virginia Police Department. She maintains an active research program and began exploring the use of computer modeling in the analysis of violent crime approximately six years ago.

She recently published a white paper outlining work with data mining and predictive analytics in crime and intelligence analysis, and has a manuscript in press in the same area with the FBI’s Law Enforcement Bulletin. She has written for and lectured to law enforcement audiences regularly at the state, local and federal level. As a direct result of this experience she can bring a unique understanding of the existing skill set and the approach necessary to convey information to this demographic.

More additional information:

McCue (2005).  Data mining and predictive analytics:  Battlespace awareness for the war on terrorism, Defense Intelligence Journal, 13, 47-63.

McCue, C., Stone, E.S. and Gooch, T.P. (2003).  Data mining and value-added analysis, FBI Law Enforcement Bulletin, 72, 1-6.