Complexity Holds Solutions to Simple Problems
Richard J. Byrne, vice president, command and control center, The MITRE Corporation, wrapped up the unclassified discussion on the first day of MILCOM 2009 by proposing that today's acquisition problems should be viewed in a different manner. Rather than thinking about how to improve what the U.S. government is doing, perhaps agencies-the U.S. Defense Department included-need to come at the problems from an entirely new direction-a very complex direction. Complexity theory can be applied in a number of areas-from acquisition to cyberthreats, Byrne explained. Emergent behavior is one of the unique aspects of complexity theory and refers to what happens when many different yet related individual items are combined. These complex systems adapt by cooperating, repeating and responding, and this results in emergent behavior, which is unpredictable. Referring to Stephen Wolfram from his book A New Kind of Science, Byrne said this means that every major complex system that exists exhibits emergent behavior; consequently, those working on complex problems or with complex networks must start thinking about them in new ways. In terms of acquisition of new systems, procurement officials must be willing to accept less than 100 percent solutions confident that products will be constantly evolving. "If you look at any of the major programs, many of them come from places such as Skunk Works. In fact, 60 percent or more of what the U.S. Army is using today in the theater of operations is not from programs of record. And yet, the dollars of the programs of record are not going into these projects. Most of the things that we're fighting the war with today are built with a fraction of the funding spent on programs of record," he related. Many contend that while these products-such as the Predator-are good individually, they do not integrate well into the systems of systems the military is trying to create. Byrnes contends, however, that once again this problem must be approached differently. Instead of looking at innovations as different, the integration issue should be looked at as different segments that reach out to each other and bleed into each other to some extent. The difference between how products from programs of record and how innovative products are built is a matter of different assumptions, Byrne explained. Innovators take into account the environment when in the design phase; if the environment the product will be deployed in is not considered from the beginning, the system will fail, he added.