The space business is becoming increasingly difficult. In terms of complexity, it is arguably in a class by itself, brain surgery notwithstanding. Test equipment can be as costly as a payload. Yet mission assurance is an absolute must.
These are the characteristics of an extremely unforgiving business. In space, it’s truly one strike and you’re out. You don’t get a second chance.
As an industry, we need to take steps to overcome the obstacles that confront us without impairing our competitiveness or attenuating the creativity of our solutions. To do so productively, however, we need to collaborate more.
By developing individual solutions to comparatively straightforward problems rather than pooling our resources in an area where rewards are relatively small, all of us suffer. T he data bases of qualification of electronic parts for use in space systems is a good example of this: All of us have this information, but we tend to be reluctant to share it.
On the other hand, if we are willing to share what we have learned from experiences all of us have been through — and none without scars — we all will benefit. Better yet, we will be able to maintain our independence. The big question facing us is: Are we ready to do so?
Basically, I see four factors hurting us as an industry in this area. First are the fiscal inhibitors resulting from acquisition in the 1990s. Second is how to effectively use COTS, or commercial-off-the-shelf, products. Third is the cost of special test equipment and resources. And finally, there is our close-hold of “industry secrets” at the lowest level. In the 1990s, NASA insisted rightly on everything “faster, better, cheaper.” The Department of Defense pushed acquisition reform. The National Reconnaissance Office demanded we maintain performance while reducing cost.
Commercial best practices were touted as the universal answer. In the late 1990s, we started cutting costs of system engineering and, occasionally, parts without fully understanding the importance of what was being cut.
The consequences of these expectations and misapprehensions were financially disastrous. They included the launch failures of three Titan 4s and two Delta 3 s. Casualties included a Mars climate orbiter, a Mars polar lander and a Mars observer. Overall, the 1990s witnessed more than $11 billion in lost space assets.
By now, the lessons of acquisition reform are well established. Bill Tosney of Aerospace Corp. said it encouraged wanton risk-taking that viewed risk as a resource. In his opinion, it eviscerated processes, specifications and standards while glorifying cost-cutting and schedule-tightening without a clear assessment of risk. It weakened technical oversight and relaxed testing discipline.
After several launch failures and many program over runs, the Young Panel found in 2002 that cost, not mission success, had become the primary consideration. This attitude bred unrealistic estimates, which led to unrealistic budgets, which ensured un-executable programs. Undisciplined system requirements seriously eroded acquisition and engineering strengths. Industry failed to implement proven management and engineering practices.
The Return to Flight Task Group reported in 2005 that mission failure was the result of erosion in four fundamental areas: rigor, risk, requirements and leadership. Standard processes were not rigorously and consistently understood or followed. Critical assessment languished. Basic understanding of requirements was lacking. Leaders failed to set the proper tone, establish achievable expectations or hold people accountable for meeting them.
As Richard Feynman noted in the Rogers Commission report, “Reality must take precedence over public relations, for nature cannot be fooled.”
Nowhere is this admonition more important than in the realm of COTS, which harbors several serious disadvantages.
One is the shrinking base of space-qualified suppliers. Another is reconciling the need for absolute, unimpeachable quality with the pressure to buy off the shelf while knowing full well that the commercial market tolerates failure. No manufacturer mourns the death of a cell phone in two years from the effects of tin whiskers.
A third drawback is that COTS products are not designed for space. Still another is that COTS folks are not known for a willingness to offer information. They are not likely to volunteer, for example, how to test and verify integrated circuits after radiation testing.
Most of the cost of space vehicles is incurred in development, whereas in aircraft, most of the cost arises during the lifecycle. Therefore, several inhibiting factors deserve mention under the heading of special test equipment and resources, which are applied to the only manufactured system that is expected to operate flawlessly for 10 to 15 years without maintenance after deployment.
Compounding that expectation is the fact that a spacecraft cannot leave orbit for repairs. So “test like you fly,” in the words of retired Lt. Gen. Brian Arnold, former commander of the U.S. Air Force Space and Missile Systems Center, becomes a critical practice in an industry where a first flight can cost more than $1 billion.
Another salient point is that test and verification requirements increase as a system grows in complexity. Somewhat contradictorily, however, testability wanes as complexity waxes. And let’s not forget that complex interfaces are a defining component of the space game.
Then there are the accounting considerations. The huge cost of test equipment must be amortized against a small production quantity.
Also we must accept the limitations of the test equipment itself. Its accuracy depends on the evolving resolution and integrity of sensor technology. At a sensor house like Raytheon, that is a major concern.
The final inhibitor is the practice of close-hold of those “industry secrets” at the lowest level. The best example is the data bases of qualification of COTS parts. We all have them but we do not share them even though NASA maintains a database of acceptable parts for spaceflight that is generally up to date. We could avoid a lot of mutual cost if we shared.
Now that I have covered what I characterized as basic reasons for the problems we face, I find myself confronted by a question for which a simple answer probably does not exist. Could we all be concentrating on the wrong issue? Despite all that I have said, the space community cannot avoid seeking faster and cheaper methods of operation. But in addition, we must deal with increasing complexity. After all, rocket science isn’t brain surgery!
The real competition is to meet the demands of our customers against a backdrop of complexity increasing rapidly in accordance with Moore’s Law, which holds that the number of transistors per square inch of integrated circuit doubles every 18 months. Thus this year’s payload may seem — and may well be — almost infinitely more capable for its mission than last year’s model.
Every space mission area is being upgraded with ever-increasing capability and integration in response to the demand for next-generation products and systems. Ever more capable unmanned probes will precede further manned missions.
Space-borne observation continues to be critical. Capability requirements rise inexorably.
We need to recognize that although we remain in competition, the contest lies in the design of ever-advancing capabilities.
We must focus on the true nature of the competitive edge and eliminate what is simply costly. We must realize that the drive for mission assurance does not weaken competitiveness. Nor does establishing mission success as job No. 1. In short, we need to collaborate on common challenges that no longer contribute to competitive advantage.
Perhaps that is the simple answer, after all.
Nick Uros is vice president for the Advanced Concepts and Technology Group of Raytheon Space and Airborne Systems.
