Big Data, Software Continue to Stump Defense Programs

Photo: Air Force

A struggling effort to upgrade Air Force combat operations centers speaks to larger problems that continue to dog information-technology programs across the Defense Department.

Buzzwords like “data fusion” and “open systems” are part of the lexicon in most big-ticket acquisitions of defense technology, but bringing that vision to life has been difficult at best.

The latest illustration of this challenge is an Air Force project to modernize command centers that are deployed in strategic parts of the world to plan and execute air warfare operations. The three-year-old program suffered a major setback last month as it became clear that the upgrades are going to take much longer and cost far more than expected.

The project, known as air operations center, or AOC 10.2 is a complex “system-of-systems” made up of least 45 different third-party software applications. The improvements are intended to give commanders modern decision-making tools, including real-time intelligence and data to make targeting faster and more accurate. That requires considerable software integration and machine-to-machine data transfer to produce more timely data and reduce human error. Making this effort even tougher are stringent cybersecurity requirements to protect highly sensitive information.

AOC 10.2 appeared to be sailing smoothly after the Pentagon signed off on a preliminary design review in 2013. After that Milestone B decision, prime contractor Northrop Grumman Corp. was awarded a new contract option to continue the development.

But after discovering significant problems over the past year, the Air Force in November submitted a “critical change” report to Congress concerning AOC 10.2. It indicated that the program — considered a “major automated information system” — would need more time and money to reach its goals, doubling development cost from the original estimate.

The critical change report was required due to a schedule delay of Milestone C — production and deployment phase — of more than one year from the original plan, said Air Force spokesman Capt. Michael Hertzog.

Air Force leaders had conducted a in-depth evaluation of the AOC 10.2 program, which led to the revised plan, Hertzog said in a statement. The AOC 10.2 needs to fix “point-to-point integration and cybersecurity shortfalls of the currently fielded version, AOC 10.1.”

Many of the troubles involved converting the legacy AOC environment into a “modular open systems architecture,” which the Pentagon now requires of all its information systems. A modular, open architecture is the Holy Grail in defense systems because it allows the Pentagon to insert new software and keep technologies up to date more easily.

“It improves the Air Force’s ability to integrate similarly modular and modernized application updates or new applications as they become available,” explained Hertzog. That also applies to cybersecurity, as the Air Force seeks to “improve system security more easily as threats continue to evolve.”

The new target for Milestone C is January 2019.

The Pentagon’s top weapons tester J. Michael Gilmore reported that major cybersecurity problems in AOC 10.2 were identified in August and September 2015. “The severity and quantity of the functional and cybersecurity deficiencies identified during the test resulted in the Air Force issuing a cure notice to the prime contractor.”

Northrop Grumman spokesman Brandon "Randy" Belote referred specific questions to the Air Force. In a statement he said the company is working to “ensure that the AOC 10.2 successfully provides for the security of the system including against future threats it will face. While there have been some challenges on the AOC 10.2 program, Northrop Grumman and the Air Force have forged a strong partnership that is working together to address the issues.”

The critical change report submitted to Congress projects the development of AOC 10.2 will cost $745 million, compared to the original $374 million estimate, Bloomberg News reported. The report said the Air Force had underestimated the complexity of integrating numerous third-party software applications and ensuring the networks were sufficiently protected from future cyber intrusions.

The Air Force Electronic Systems Center at Hanscom Air Force Base, Massachusetts, first issued a request for proposals in December 2010 for the AOC upgrade, estimating the value of the program at more than $800 million over the next eight years.

The modernization of air operations centers has been a long-term pursuit by the Air Force, noted retired Gen. Charles F. (Chuck) Wald, a former air war commander and now vice chairman of Deloitte Services.

The integration and interoperability of equipment at the AOC has “always been an issue,” Wald told National Defense. “How do you make sure you have feeds from the various intelligence sources and monitoring sources, and how do you apply that capability to an air tasking order?” he said. Open architectures and data fusion, if executed properly, are “game changers.”

During an industry conference more than a year ago, Deputy Assistant Secretary of Defense for Systems Engineering Stephen P. Welby spoke about the difficulties in developing modular open systems. One of the obstacles is a lack of “technical insight” by government program officials. “These designs will increase demand on DoD engineering competence, capability and capacity.”

Vice Chairman of the Joint Chiefs of Staff Gen. Paul Selva recently called out contractors for not being forthcoming about the challenges of building open systems. “For those of you in industry that are in this room, I can't tell you how many times I've asked the following question, ‘Will your widget subscribe to an open architecture?’ Answer is always, ‘Oh sir, of course,’” Selva said in October at the Center for Strategic and International Studies. In reality, though, it’s an “open architecture but only inside of our company or only inside of our proprietary IT that's in the system,” he said. “We really have to find a resilient open architecture to which all of our systems can subscribe and we've only scratched the surface on that.”

The Defense Department’s top weapons buyer Frank Kendall said the idea of modular designs and open systems goes back decades. “It's been part of my initiation since day one,” he said. “How do you basically keep those systems modern? Well, you do it through modular designs. Modular design gets you the ability to take something out, put something else in,” Kendall said. “Industry always tells us that they like open systems, but they give us a lot of designs that aren't open. There's no secret about this. You basically want to retain market share and one way you do that is you have proprietary intellectual property that allows you to do that. It makes hard for people to come in and displace you.”

The Pentagon has to “work hard at this and the devil is in the details,” Kendall added. “That's the only way we're going to have technology refresh on reasonable cycles relative to the pace at which technology is moving.”

Topics: Infotech, Defense Department