Prepared for liftoff

The Air Force's 45th Space Wing is developing a knowledge management framework to simplify sharing lanuch data

Nothing about sending a rocket into space is easy.

Many players are involved'civilian and military agencies, contractors, commercial customers. The work it takes to safely launch a rocket and its payload goes on for months, sometimes years.

Coordinating data and sharing it among the participants has always been paper-based, labor-intensive and time-consuming, punctuated by many sit-down meetings to make sure everyone was on the same page.

But a small project that is being conducted under the auspices of the Air Force Research Laboratory is trying to create a knowledge management framework that streamlines part of the work, with the potential to provide a universal tool for most, if not all, stakeholders.

The potential payoff is high. Richard Sirmons and Richard Thiebauth of the Air Force 45th Space Wing, the in-house champions of this knowledge management initiative, say a 6-to-1 payoff for the investment would be a conservative estimate.

Working with AFRL and a Florida-based company that holds two Small Business Innovation Research contracts, Sirmons and Thiebauth are directing the creation of what they call a federated database for a virtual enterprise.

The 45th Space Wing, based at Patrick Air Force Base in Florida, is responsible for the Eastern Range, Sirmons said. This comprises instrumentation sites scattered throughout roughly 15 million square miles that form a circle with Cape Canaveral as the hub, stretching from Ascension Air Station in the South Atlantic Ocean to Nova Scotia, Canada.

'We coordinate all the support functions for launch vehicles,' Sirmons said. 'Delta II, Delta IV, Atlas V, the space shuttle. We just launched the last Titan.'

But the wing doesn't 'own' the sites, nor does it generate the vast majority of data that has to be compiled, he said.

The launches can be carried out on behalf of the Air Force, other Defense Department agencies, civil agencies or commercial interests, even submarine launches.

Those entities all have to coordinate with their own contractors and subcontractors, then with each other, making the tracking and allocation of assets particularly complicated.

The wing provides support for several major contracts, Sirmons said. Among them are the Range Technical Services Contract for operations and maintenance on radar, optics and communications; the Launch Operations and Support Contract for maintaining launch facilities, launch pads and processing facilities; and the Joint Base Operations Support Contract, a combined contract with NASA at the Kennedy Space Center to provide security and police, fire protection and noncritical security functions.

'This is all the data that supports the various processes that support preparations for launch, then the launch itself,' he said.

The question has been how to get information back from semi-autonomous agencies and contractors, Sirmons said. The goal is to build an enterprise model, which he calls an ontology, that accommodates the needs and interests of all parties.

'We're after the five rights: getting the right information in the right format at the right time for the right person to make the right decision,' he said.

The SBIR program plans to award fairly small contracts to small companies to undertake research and development in areas specifically of interest to the government.

In this case, said Richard Hyle, program manager at AFRL, a small company called Modus Operandi Inc. is taking current technologies for data sharing, doing research about how to make them more effective, and using the 45th Space Wing's data-sharing problem as a real-world test case for its solutions.

'This will lay the foundation for effective information sharing,' Hyle said.

'The knowledge management initiative is more than the technology. It's processes, teaming,' said Peter Dyson, CEO of Modus Operandi of Melbourne, Fla. The company is the solution provider for the KM framework, he said, while Sirmons and Thiebauth are responsible for convincing all the parties to come together in a loose community of interest.

The company received two Phase 2 contracts about a year ago, each for two years and worth $750,000, to conduct the work.

'From the standpoint of the SBIR program, we're at the midpoint,' Dyson said. Deliverables typically are due at the end of a contract, 'but we structured the program very differently, where we've actually already delivered two versions of the system, with latest version delivered in January.'

The company is pulling information from three databases for the first application, the sharing of scheduling information for all kinds of assets, including physical, human resources and technological.

'The intention is to continue to add information sources over time,' Dyson said.

Thiebauth calls the knowledge management framework a federated approach because there is no central repository for the data. Instead, each group keeps custody of its own data but makes it available to other stakeholders.

'This is not a monolithic control system,' he said. 'We build incrementally, which builds trust.'
The team did a business case analysis of a typical scheduling effort, Sirmons said, and found 'it was reduced by a factor of six to one,' a significant return.

But one drawback, Thiebauth pointed out, is that it uncovers the reliability'or lack thereof'of the data being shared.

'There can be concerns about data integrity,' he said. 'There are a variety of problems. Some data is just wrong. You need authoritative sources. Then there's the currency of data. The metadata provides those descriptions.'

While Sirmons, Thiebauth and others continue to build their community, Hyle at AFRL looks ahead to the final step.

'Phase 1 is basic research, Phase 2 is application to real world problems.' The third phase is the commercialization of the product, which grows out of the fact that the products are not designed to be domain-specific, but to have much broader applicability.

The small businesses that benefited from the contracts to develop real-world applications are then free to find their own funding to take the new, already tested products to market.

inside gcn

  • artificial intelligence (ktsdesign/Shutterstock.com)

    Machine learning with limited data

Reader Comments

Please post your comments here. Comments are moderated, so they may not appear immediately after submitting. We will not post comments that we consider abusive or off-topic.

Please type the letters/numbers you see above

More from 1105 Public Sector Media Group