Military aircraft today are expected to endure far beyond their originally planned life spans: the 41-year-old B-52s may be around for decades to come. Military airplanes also have to accommodate new weapons, deal with new threats, respond to new mandates, and adjust to the rapidly changing electronics environment. Today’s avionics systems were not designed for this.
Driven by commercial needs, electronics components and techniques, along with software programs, architectures and languages, evolve quickly. With less than 1 percent of the electronics market, the military can’t influence this trend. New aircraft become obsolete before they are fielded: the F-22 will have been through four electronics upgrade cycles before the first production aircraft rolls off the line, according to a recent National Research Council (NRC) study.
The cost of the aging avionics problem is huge and growing, while the funding allocated to fix it is decreasing, warns the NRC study, citing FY2001 budget documents. The Air Force spent about $1 billion on depot-level maintenance and support of avionics systems in FY1999. The report estimates these costs will increase about 50 percent by FY2005. The total cost "is not well-known because the accounting process is not geared to accumulate [that] information" says Noel Longuemare, vice chair of the NRC study and former principal deputy undersecretary of Defense for acquisition and technology. He believes that funding for the problem is highly uncertain.
The problem impacts readiness, too. In the 1990s, the aircraft mission capability rate declined from 83 to 73 percent. This reduction is "due largely to the increasing age of the aircraft fleet, particularly the aging avionics systems," argues the NRC report. The three services face a "dual problem," says Fred Ziska, customer advocate-upper midwest, for Rockwell Collins. "Mission capability rates decrease every year, while maintenance costs increase. You’ve got two ‘death spirals’ to get a handle on."
MOSA the Answer?
The Air Force last year established an Aging Aircraft Program Office to advise on avionics and other issues. There’s also a "need for top-down direction" at the Defense Department level, Longuemare says. Key elements of such an enterprise strategy, he says, would include:
A modular open system approach (MOSA) mandate,
Roadmaps for technology insertions, and
Funding for the roadmaps.
MOSA is a "set of concepts that drive affordability over the life cycle," explains Rich Eisenhart, senior director of integration applications engineering with Rockwell Collins. "MOSA doesn’t define an architecture, but a framework of standards and reference models." Integrated change roadmaps, as recommended by the Air Force Materiel Command, identify avionics viability issues and integrate that planning process with existing upgrade programs, says Will Urschel, avionics technical advisor for the Aging Aircraft office. About 20 weapon systems developed initial roadmaps last year.
However, the problem is, "we don’t see a concomitant funding stream" for the roadmaps, Longuemare says. "Quite frankly, the plans are meaningless if there’s not money put in the budget to implement them." Funding is tight, Urschel acknowledges, but "the objective is to leverage funding already targeted at sustainment and capability improvements to provide incremental [viability] improvements, as well."
Long-term analysis is useful. A study in the early 1990s of alternatives for upgrading the T-38 advanced flight trainer helped to popularize a process model that incorporates issues such as an avionics migration strategy, open systems considerations, and life-cycle cost analysis in upgrade planning. The study indicated that a major avionics modification would have about the same total cost as a "band-aid" approach, says USAF Lt. Col. Glen Logan. Logan, now deputy, systems integration, with the Open Systems Joint Task Force, conducted the study. Operations and support (O&S) costs for the bare-bones mod were projected to be more than three times the cost of the more extensive solution. The T-38C program chose a path midway between these options.
The aerospace industry already is adopting modular, open systems approaches to avionics. Boeing, Lockheed Martin, Rockwell Collins and Northrop Grumman have such efforts. But "right now, it’s an ad hoc, sort of helter-skelter thing," Longuemare contends. "Many programs are not even aware it’s important."
But the Aging Aircraft office cautions against pushing MOSA as a requirement. "We don’t even use the word, ‘open,’" Urschel says. "We use ‘viable.’ Viability means looking at how you’re planning on using your product over time and assuring it meets your needs and is affordable, he says. "There is no generic architecture I’m going to ask for." Rather, the emphasis is on goals such long-term upgradability and testability.
Past and Present
Many avionics systems were designed to allow 100 percent growth in initial capability–memory, bus, processing capacity–over a 20- to 30-year life span, explains Urschel. "But instead of 2 to 3 percent growth per year, we’ve seen a 10 percent rate. That meant that [planned growth] would be saturated in seven to 10 years, making any additional capability much more expensive." To squeeze out more performance, "you would have to go back in and basically redesign a lot of the existing system," Urschel says. In the past "the Air Force identified a design implementation and directed it to our contractors," he says. But now the emphasis has shifted to systems that can adapt to change and whose changes are easily verified. "If we tell [bidders], in a gross sense, what our capability upgrade plans are, they can forecast what the technology updates will be and [identify] components with high turnover risk."
"In the past, we believed if we could dictate a design solution and make it common across platforms, that would give us a cheaper system. I’d rather that each of the products be common with a commercial product line, where they’re leveraging off of millions on the commercial side."
The Aging Aircraft office has developed contracting tools to reward companies that propose long-term performance approaches despite upfront cost penalties. The office has devised a "viability index" formula for the consistent, qualitative assessment of business practices, processes and system architectures vis-à-vis long-term affordability, capability and availability goals.
The C-130 Avionics Modernization Program (AMP) is trying some of these strategies. Bidders were asked to complete a "what-if" study, hypothesizing the need to add extra capability in five to eight years. Respondees showed how their design methodologies, architectures and business practices would rise to the challenge.
There was nothing binding in the contract language, Urschel says. But the "cost plus award fee" contract type–which assesses contractor profit annually–could provide a useful stimulus. Boeing has developed a long-term viability plan, he says. That plan and the contractor’s execution of it could be major factors in assessing profit–and spurring viability performance.
Viewpoint: Aging Aircraft Program Office
Recently, Avionics Magazine interviewed Brig. Gen. Rosanne Bailey, director of the Aging Aircraft Program Office at Wright-Patterson Air Force Base, Ohio. Her office advises the Air Force leadership on issues, such as aging avionics and aircraft structures, which extend across the aeronautical domain. The unit will soon become the Aeronautical Enterprise Program Office, a move whose significance she explains below.
Avionics Magazine: Could you explain the aging avionics problem?
Bailey: In avionics the technology changes so fast. But with budget cycles, you have to plan your money three years away. So, in one six- to eight-year development cycle, the technology may turn over eight times. It’s not just components; it’s architectures, approaches. You might have a hardware turnover and then a spectrum turnover and then a software change. It’s all the different aspects of technologies that affect us.
And each technology has a tendency to drive the rest of them into the next developmental cycle. They’re all flowing, one right after another. So we end up with systems that are non-viable, from a parts obsolescence point of view, before they even get produced.
We’ve tried "stopping" change–freezing a configuration–contracting with some company who will agree, for all time, to produce these parts. Well, it’s just not possible after a while. If a company wants to compete in industry, it cannot maintain one foot locked in the past, the Neanderthal Age. Antiquity is two years ago on some of our computer systems.
Avionics: Can you size the aging aircraft problem in dollars?
Bailey: I’m not sure I can quantify the dollar size of the problem because we are still discovering it. It’s a huge problem, and the money we have against it is not enough. It’s getting more expensive, and even the…Congressional Budget Office [CBO] analysis said [the cost is] going up "only" 10 percent per year–which, compounded per year, is outrageously expensive.
The thing about the dollar size of the aging aircraft problem is that it depends on what you want to do with [the aircraft], which changes regularly. It depends on what option you select to solve which problem–we’re constantly in motion on that. And then, we’re constantly affecting it with things we’ve already started, so it’s moving all the time. What is clear, though, is that, without intervention, the trends are upward for the cost of repair and maintenance on every single platform. Our task is to bring the trend line down, which will buy us time to field new solutions.
Avionics: How about your own budget? Is it going up?
Bailey: The budget doesn’t feel like it’s going up to me. But you have to realize that the majority of our budgets are going toward the need for current conflicts, as well as already programmed things. The budget last year was about $30 million for all our aging aircraft and technology projects. If we’re lucky, we’ll stick pretty close to where we are now.
Avionics: Your organization is still evolving?
Bailey: The organization is about to become the Aeronautical Enterprise Program Office. In a nutshell, its new role will be to connect all the current elements of the aeronautical world, or enterprise, to ensure we provide a complete and integrated set of options to decision makers, and then to connect aeronautical to our other three enterprises: armament, command and control, and space. In other words, in Aging Aircraft we work to keep ‘em flying. In Enterprise, we work to also keep em’ relevant for the Air Force of the future.
Avionics: Are some platforms at the upgrade saturation point?
Bailey: I don’t think there is such a thing as a saturation point, given the continuing miniaturization of systems. But sometimes you may have to upgrade power, for example. Or you may have to upgrade data pipelines in a system. In other words, you may have to upgrade more than just a box in order to get the greater capability. In other cases, some capabilities simply cannot be retrofitted. We cannot take a platform such as our F-15s, which otherwise are great aircraft, and retrofit stealthiness into them. That’s a primary design characteristic that must be considered in the initial design of the system.
Avionics: Do you think the much-touted MOSA [modular open systems approach] is the answer to the aging avionics problem?
Bailey: MOSA is one approach to the problem. But it is another case of directing a solution rather than a performance attribute. MOSA may be the perfect solution in some cases, but it might not be. I don’t want to lock in today’s solution. It’s taken us a long time–but we think we’ve got it now–to quit dictating solutions and instead dictate the performance attributes we want to see, such as affordability, ease of upgrading, and testability. [Setting performance requirements] means I can update easily, fast; I can update for less cost; and I can port my existing software with very little change in software. Viable Combat Avionics [a major thrust in the Aging Aircraft office] is about dictating performance attributes –not the architecture, but an architecture which enables change.
One characteristic of an architecture that enables change…is separation of hardware from software. For example, you can take [Microsoft] Office, move the computer out from under it, put a new computer in and it still works.
Avionics: How do you hope to infuse these ideas into the system?
Bailey: We’re not going to interfere in the [program management] execution chain. But we’re going to provide something of a consulting service, if you will, and a knowledge management service.
We’ve had a tendency to identify lessons over and over. But lessons are only learned when you apply them to something. We’re not very good at doing that…because the people who need the information don’t know it’s there to be had. Typically, the only way you really transfer lessons is if someone who knows what you need to know happens to transfer into your organization. That’s a little haphazard.
Avionics: How are you going to ensure that people pick up the message?
Bailey: That’s the part about making the performance attributes–upgradability, affordability and testability–major requirements at the top of the chain. Our goal is to get agreement that these attributes are requirements vital to the long-range viability of all of our systems. If they become threshold requirements for a system to be judged successful and…able to go into production, then the program managers will come looking for how to do this. So we’re going back to those who write the requirements in order to make sure it gets in there…for any update to any program, as well as for any new programs.
Avionics: Your office has written some procurement contract language?
Bailey: We have written some language for use in solicitations, contracts and award fee programs, as well as revising our best value methodology for use in source selection. This enables us to ask for the right thing, evaluate it fairly, and reward the contractor who comes in with a proposal that may cost more in the instant contract but will reduce the cost over the life of the system, well beyond the contract life. By doing so, we will communicate to industry that we have the will, the real intent, to make this real and that, in fact, they’re not going to get priced out of the market by those who come in at lower cost but without a plan for the future.
I believe the dollars will come when people believe you’re going to do something useful. We also have to show some successes. I think that when the C-130 [Avionics Modernization Program] and the C-17 [programs that are working with the Aging Aircraft office] start to show it’s working for them, it will make a difference in how Viable Combat Avionics is viewed.
Rockwell Collins’ CORE
Rockwell Collins’ Common Reusable Elements (CORE) approach–begun in 1994–touts more than modularity and openness: it adds software partitioning, to allow one software component to be changed without affecting others; connectivity, so that processors can access the information they need; and supportability.
The CORE architecture applies across the enterprise–military, business/regional aviation and air transport, says Rich Eisenhart, senior director of integration applications engineering. "CORE defined a set of standards and common reference models to drive [component] reuse."
As prime contractor on the KC-135 Global Air Traffic Management (GATM) upgrade program, Collins applied CORE reusability principles. The KC-135 uses Collins’ FMS 6000 flight management function from the company’s commercial side. (FMS 6000 functions are hosted as a partition of Collins’ virtual machine operating system, VMOS–itself a version of the commercial LynxOS operating system.) Because the commercial FMS software wasn’t designed to handle military operational refueling patterns, Collins created another software partition for that function. Isolating civil from military functions is intended to make upgrades easier.
Software partitioning also "saves millions on recertification testing," says Fred Ziska, Collins’ customer advocate-upper midwest. "You only have to recertify the portion of the system that’s been affected."
The CORE approach is efficient because it leverages Collins’ three aviation markets. "We don’t have to pass back all those costs to the military any more," Eisenhart says. "The civil side pays for upgrades that deal with new [civil] requirements. We’ll be able to lift those [upgrades] up directly and reapply them for a very affordable price."
A final touch: as hardware and software modules exit the Collins factory, they are "taxed" to support the development of product line roadmaps, Ziska says. The Aging Aircraft office had not heard that from any other company, he asserts.
Boeing’s Bold Stroke
Bold Stroke, Boeing’s implementation of open system, modular concepts–begun in the mid-1990s–aims to reduce total ownership cost (TOC) by 50 percent, compared with historical levels, says Don Winter, director of open systems architecture at the Phantom Works Division. Boeing has applied the approach to the AV-8B, F-15, F/A-18 and T-45. Data indicates that development costs on applicable AV-8B, F-15 and F/A-18 programs have met or exceeded the 50 percent goal, Boeing claims. (TOC includes components such as initial hardware and software development, software upgrade and maintenance, and hardware sustainment costs.)
Bold Stroke encourages reuse of processing and software components. The F-15 and the F/A-18 boast a common image processing module. Navigation software components are shared by the AV-8B, the F/A-18, the F-15 and the T-45. And, thanks to reuse of software components from Bold Stroke and other programs, the T-45, a newer program, took 80 percent less time to develop functional aircraft mission software than typically has been the case, Winter says.
Bold Stroke also emphasizes openness–the use of commercial operating systems and software tools, programming languages, mainstream techniques, such as object-oriented design, middleware standards and interconnect protocols. "Well-defined interfaces, based on commercial standards, can remain stable" in the midst of change, Winter explains. More abstract principles include hardware/software isolation and "change containment" via modularity.
Boeing aims not only at cross-platform, "horizontal reuse" of software components but at "vertical reuse" on simulators and developer desktops. Because flight software modules are developed to be portable, flight code can be used in simulators, a major savings. And flight software can be run and tested on a developer’s desktop using commercial operating systems such as Windows NT, VxWorks, Solaris, LynxOS or Linux.
Bold Stroke is also a business model, Winter says. It redefined prime/supplier relationships and changed the business models of some Tier 1 suppliers, who now "assemble systems from commodity components." The open system approach has had a "significant impact on suppliers’ bottom lines," he adds.
Boeing has struck some teaming relationships with suppliers to encourage them to "co-invest," Winter says. Honeywell, for example, a Bold Stroke partner from the beginning, has been involved as a supplier of display and video processing modules on the F/A-18 and the F-15. The partnership "opens up opportunities in the future to get onboard other programs," Winter says.