The Pentagon’s Inspector General (IG) released a report yesterday about the U.S. Ground Based Midcourse (GMD) missile defense system: “Exoatmospheric Kill Vehicle Quality Assurance and Reliability Assessment, Part A.” It’s telling.
Part A assesses the quality-assurance processes that Raytheon, the subcontractor, and Boeing, the prime contractor, used to build the missile defense system’s Exoatmospheric Kill Vehicles (EKV).
Part B, which would be really interesting, looks at the reliability of the deployed EKVs, and is classified.
Forty pages discussing defense contractor quality control isn’t the most riveting reading, but there are a few things that are pretty interesting.
How did a missing piece of wire cause a $200 million failure and set the program back by a year?
The Missile Defense Agency (MDA) said that the January 31, 2010 FTG-06 missile defense test failed because “a lockwire was not inserted during the EKV manufacturing process.” This omission caused the failure of a maneuvering thruster on the kill vehicle, which caused the missed intercept. That was an expensive quality-control mistake, and not only because missile defense tests cost $200 million or more to stage. Had this mistake not been made, the underlying design flaw in the GMD kill vehicles that was uncovered when it caused the failure of the re-do test 12 months later, might have been caught earlier. So it set back development at least a year.
What did the IG have to say about this? Somewhat surprisingly, the IG comes down pretty lightly and had not so much to say about this incident specifically, though the report says “Work instructions were updated as part of the corrective action to ensure verification of lockwire connectors.” (Were they not in the instructions before?) The IG’s comments about the larger context, however, are telling:
“The current EKV [kill vehicle] design is the prototype design of 1998 with upgrades for design and manufacturing defects, and obsolescence issues. The immediate need for an initial capability drove an accelerated development process and fielded capability before EKV performance was fully characterized prior to initial fielding. Requirements were viewed as “goals” with little focus on reliability, producibility, and maintainability requirements, which are integral to strategic systems with a life expectancy similar to GMD.
“A combination of cost constraints and failure-driven program restructures has kept the program in a state of change. Schedule and cost priorities drove a culture of “Use-As-Is” leaving the EKV as a manufacturing challenge. With more than 1,800 unique parts, 10,000 pages of work instructions, and 130,000 process steps for the current configuration, EKV repairs and refurbishments are considered by the Program to be costly and problematic and make the EKV susceptible to quality assurance failures.”
In other words, putting together a kill vehicle is a long, painstaking process prone to error and it sounds like everyone is pretty irritated by it. And this complexity is at least partially due to time and budget constraints.
Though two things come to mind. The first is that the breakneck pace to meet the “immediate need” declared in 2002 has not been seriously re-evaluated in the interim. And second, all programs have budget and time constraints, but what is particularly notable about this kill vehicle program is that large numbers of them were produced and deployed before the kinks got worked out. In fact, interceptors were fielded before the first intercept tests even occurred.
While undisciplined industrial processes in a system that needs to work the first time cannot be tolerated, the way the system is set up seems to make failure inevitable. This presumably could be addressed in the planned EKV re-design process—but only if it can be done with integrity, appropriate oversight, and without undue time pressures.
Each kill vehicle is a special snowflake
The report was also frank about poor record keeping by the contractors. Boeing and Raytheon did not reliably keep track of and communicate modifications and substitutions for subcomponents of the kill vehicles, compromising the flow of information between and them and to the MDA.
The IG writes that without proper documentation:
“nonconformances may circumvent the engineering review board and material review board processes. Approval of deviations ensures that MDA is aware of any potential impacts to the system that could result from incorporation of hardware or software that is not produced to specifications.” The lack of discipline in this area “can result in the fielding of an unapproved configuration.”
“Fielding of an unapproved configuration” sounds kind of like a high school dress code violation, but actually is critically important. Especially so since there are a mix not only of the different variants of the kill vehicles (CE-I and CE-II), but a number of other subconfigurations “that resulted from resolving design or manufacturing risks.” In fact, it’s even worse than that, as David Wilman noted in his June piece in the LA Times:
“Because each of the kill vehicles is handmade, no two are identical. A fix that works with one interceptor might not solve problems with others.”
The IG’s point about poor record keeping begs the question whether there is sufficient accurate information about each of these painstakingly assembled, sophisticated products to know how each kill vehicle will perform.
This also means that the success or failure of one idiosyncratic interceptor doesn’t provide straightforward information about the reliability of the next one. This makes ascertaining the reliability and effectiveness of the GMD system via a test program quite challenging.
The upshot of the IG report
Despite all this, following the successful intercept test in June, MDA seems to be ready to resume production of the kill vehicles. This decision means big money for Raytheon. In 2014, Raytheon stands ready to assemble 9 CE-II kill vehicles that are “in pieces and parts” in a Tucson, AZ, facility. This will recoup Raytheon $150 million in withheld payments. And then the company will assemble 8 more in 2015 and 9 more in 2016.
The Inspector General’s report describes the symptoms, but does not diagnose the underlying problem.
These failures of basic industrial quality control are not isolated issues in an otherwise well-functioning research and development program. The missile defense system is plagued with oversight and accountability issues, and operates under a “rush to failure” paradigm. The Missile Defense Agency seems to be trying to take responsibility for turning this around, but that is unlikely to be enough unless we as citizens who pay for missile defense and who are meant to be the beneficiaries of it demand accountability. We as citizens—and Congress— should be asking harder questions: What’s the rush and who is benefitting from it? Is this the best way to spend $40 billion? Is this the best way to manage nuclear threats?