Big obstacles remain for big data analytics at DOD
- By Mark Pomerleau
- May 16, 2016
Despite calls to integrate big data analytics into everyday military operations to make sense of vast sets information, it might not be the panacea some advocates claim. And while service branch IT executives would be among the first to say that analytics are important, they have their reservations.
Before investing in big data analytics, the military should evaluate the effectiveness of the analytics it currently produces. “We have a lot of analytic tools. We haven’t gone back to look at whether or not the algorithms or what they produced are actually accurate,” Brig. Gen. Dennis Crall, director C4/CIO for the Marines Corps, said. “We also haven’t taken the time … to look at the quality [of the data we have] or the source data that feeds it.”
The first step should be ensuring data comes from consistent, reliable sources. The information regarding the number of Marines are on active duty, for example, is “widely disparate,” Crall said at a May 11 AFCEA-hosted event. “No one looks at the source stream for these things, but we’re trusting what comes out the other end. So ensuring that we have discipline in the architectural pieces for these analytic tools is critical.”
Data quality is the second step. “We have found that even if we got the sourcing right, we don’t have the discipline in the services -- in our own service anyways -- to ensure that those who are doing the data entry are doing it right,” Crall said. Data entry errors can significantly impair the best algorithms and lead to inaccurate analysis.
Military leaders also must understand what a particular data analytics solution can actually provide., “You really have to understand the value proposition,” Army Deputy CIO Gary Wang said. “Is the value proposition going to be monetary or is it going to influence or [result] in a better decision being made?”
It’s one thing to talk about data-driven decision making, but it’s important to understand that data is the foundation of the digital value chain, he said. Next, information is created from the massaging of that data, with knowledge or insight being the ultimate goal.
Is “analytics going to help you move up that value chain from data to information to knowledge," Wang asked, "and is the insight going to pay off in terms of a decision?”
“If the idea is to really push analytics, it’s important to be able to point to a definitive cost-saving measure,” a decision that saved a life or prevented a suicide, Wang said, noting that social media and analytics could help the military with suicide prevention.
Smart metering is another untapped area in which analytics could deliver insights. Data analytics is often applied to “IT infrastructure, but gas, power, water -- those are all important to the military. Smart metering and analytics can really help on that,” Wang said.
Mark Pomerleau is a former editorial fellow with GCN and Defense Systems.