There are at least two approaches to cutting through the glare. One can increase the speed and effectiveness of processing to keep up with the volume of information being collected, or one can collect the information more selectively, reducing the need for processing. The first approach typically adds processing power and human analysts to handle very broad sweeps of information. The second depends upon knowing what kind of information to go after and where to look for it.
Recent advances in geospatial information and knowledge discovery are making more selective collection increasingly attractive. We’re able to make better use of our finite search and surveillance resources, and we’re able to make more discriminating use of powerful collection tools that could, unchecked, overwhelm our processing and analytical resources.
The emerging discipline
With the emerging discipline of predictive spatial intelligence, we are seeing these capabilities come to fruition. Using robust computer platforms, we can now combine geospatial information from multiple sources and analyze and synthesize the data to predict the presence of items of interest without setting one foot on the ground. We take a well-studied geographical location, find patterns in its known phenomena and then apply those patterns to unknown regions to predict the presence or absence of like phenomena. We decrease manpower needs by improving the capabilities of analysts and using predictive spatial intelligence to increase effectiveness of resource deployment. This kind of integrated information approach is a flexible way to use and reuse information gathered in traditional or non-traditional ways. For missions that require location of items of interest, we can use this approach to determine the areas likeliest to contain the items, allowing us to pinpoint our further efforts and manpower.
Conceptually, this is not entirely new: The old military disciplines of terrain analysis, intelligence preparation of the battlespace and operations analysis were all fine-tuned to produce smarter searches and more relevant information. What is new is the confidence with which predictive spatial intelligence enables us to focus our attention. As information technology advances, the capabilities predictive spatial intelligence provides will expand and improve. Funding invested in the development of this technology will see a return of investment and reduce the risk associated with resource deployment by narrowing the scope of deployments to areas likeliest to be successful.
This technology takes advantage not only of electronic information but personal knowledge, as well. Experts can create models than can be combined with technological models to bring, for example, new depth to search missions. When analysts and experts turn their knowledge into data sets that can be saved and shared, not only does it benefit the mission at hand but also future missions. When knowledgeable analysts retire or are reassigned, the information they contributed is retained. With digitized expert knowledge, predictive spatial intelligence can inform decision-making with all available information, thus increasing the likelihood of success.
Surveillance, like search, is expensive and labor intensive. But surveillance educated by predictive spatial intelligence can yield faster results at lower costs. By keeping track of past incidences, it is possible to shape a surveillance effort based on accumulated knowledge of an area. Do the facts on the ground point to a better way to execute surveillance? Is there a better location at which a surveillance effort could be headquartered? These questions may not even be considered if the information collected is never integrated and spatially analyzed.
Predicting the future
Predictive spatial intelligence earns its name because of its potential to predict future events based on knowledge of the past. The predictions can be used to improve logistics (think of traffic flow or supply-chain planning), to stop threats before they materialize or improve the precision of critical missions by focusing resources where they are most necessary or most likely to succeed. We already have a great deal of information, but now we need to use it to the fullest extent. More data sets and methods will further enrich an already successful composite predictive model.
Reducing cost, manpower and time to accomplish tasks is especially important in the context of homeland security—every dollar and man-hour saved can be applied to other missions. Homeland security is rightly an expansive mission, and there is no shortage of things to do. Using predictive spatial intelligence, we can leverage our information technology and the information we gather to make search, surveillance and inspection faster and more effective.
Jeffrey A. Goldman works as director of technology in QinetiQ North America’s Federal, Civil & Intelligence Business Unit, IT Services Group. He has more than 15 years’ experience in data mining, remote sensing and knowledge management technologies for defense and security applications. He earned his PhD at the University of California in Los Angeles for work on knowledge discovery in free-form media.