Over the years, the U.S. Air Force has come to rely more and more on industry for its technical understanding of the systems it acquires. To reverse that process and bring costs and risks more under its control, the service is experimenting with creating and carrying digital system models through the entire life cycle of a program.

The acquisition process now being piloted has two key elements: the “digital thread,” a model of the system that begins at conceptualization and evolves through its life to retirement; and the “digital twin,” a model of the system as built, including any manufacturing discrepancies, that is used to support the system in service.

The Air Force has tried to go this way before, in the late 1990s, when it began looking at model-based acquisition. This was intended to use modeling and simulation tools to build a digital system representation that would be carried forward into acquisition. The system would be built against this, which would be corrected as the system was built, and then used as a logistical support tool throughout its service life.

The effort ended in the early 2000s, but is coming back as part of an imperative to “own the technical baseline” of future systems, says David Walker, deputy assistant secretary of the Air Force for science, technology and engineering. “This is not about owning the technical data. It is about having access to and being able to exercise the data. We really need to understand the technical risk,” he told the American Institute of Aeronautics and Astronautics SciTech 2015 conference in Orlando, Florida, this month.

The Air Force “abdicated” its technical understanding to industry under the total system performance responsibility approach to acquisition, Walker says. “We gave industry a performance specification and didn’t care how they got there. We need to return to where we understand our systems, and digital models are one of the tools that enable that.” Digital-system modeling is used in various areas within the Air Force, but the effort now is to make it a consistent thread throughout the acquisition life cycle of a system.

The Air Force has three pilot programs under way to apply digital-system modeling to different stages of the acquisition process and, based on the results of these, intends to use the digital thread and digital twin to “own the technical baseline” on four upcoming procurement programs.

The first pilot involves the intelligence, surveillance and reconnaissance multi-resolution analyzer (ISR MRA), a tool used prior to a Milestone A decision to begin technology development to understand how the new system fits into an ISR architecture. The initial pilot will use MRA to develop a capability within the Air Force’s Simulation & Analysis Facility to test new ideas using live, virtual and constrictive simulation “so we can fly before we ever buy anything,” he says.

The second pilot will use the Create AV computational tools developed by the government to enable assessment and analysis of air vehicle designs. “We have a system that is already developed and a digital model that is already developed. Using those, can I go back and predict the outcome of wind-tunnel tests? We know the reality, because we have the wind-tunnel database, and we have the complete system model,” Walker says. The goal is compare the two, and see how wind-tunnel testing can be reduced to where it is used only to verify the model.

The third pilot involves manufacturing and the material review board process now used to determine if a non-conforming part meets requirements. Today, if a part passes that review it loses its non-conforming label and goes into the system as a normal part, he explains. The pilot “will build a digital twin of that part to understand how it impacts the system in the long term.” This is a first step toward a digital twin on a complete aircraft.

The four planned acquisition programs to which the pilot results will be applied are recapitalization of the Joint Stars battlefield-surveillance fleet (see artist’s concept); the T-X replacement for the T-38 advanced trainer; the OCX operational control system for GPS satellites; and the Long-Range Stand-Off (LRSO) replacement for the nuclear Air-Launched Cruise Missile.

“For Joint Stars recap we will work a digital twin as we go into Milestone B [development go-ahead] and through manufacture. We will build a digital twin for each aircraft off the line. It will be the first real test [of the process],” says Walker. “T-X will be different, as it will most likely involve building the digital thread for a commercial off-the-shelf system.” For LRSO, the Air Force is to “take a different approach to design of the missile, so we need a good understanding of manufacturing up front. We will need good digital models as we go into Milestone C [production decision], so we know what makes a good missile.”