The top defense industrial issue for the next administration -- whether one overseen by McCain or Obama -- is to improve the "very poor record" of planning for and building the country's weapons, says Paul Kaminski, one of the wisest minds in the business who also happen to be an advisor to Sen. Barack Obama's (D-Ill.) campaign. Just to be sure we are all on the same page, Kaminski stressed he was speaking as a former undersecretary of defense for acquisition and technology and not in his capacity as an advisor to Obama,
"The top issue for both government and industry has to do with our ability to execute on acquisition programs, where we have had a very poor record of late," he told me Tuesday. The keys to restoring competent execution and shortening the ridiculous amounts of time now needed to develop and field most major weapon systems are greatly improved systems engineering, combined with heavy reliance on prototyping, at the beginning of each program. That must be supplemented with much greater attention to the building of all-important subsystems, upon which every major program relies.
He points to a program with which he is very familiar, having overseen it while at the Pentagon, modernization of the ICBM. The program hired TRW to perform systems engineering and the company agreed to forego work on development or production "so they could participate in an honest way," Kaminski says.
Another key ingredient was "something called development planning" that was used for the program, Kaminski said. It helped the ICBM program make sure that its major subsystems did what they were supposed to do and could integrated on the platform at the right time so the program did not fall behind due to the weakness of a subcontractor.
Doing these things in the early stages of a program should help reduce risk and speed development since much less work will be done fixing subsystems once they have already been integrated.
"I come from a period when we were doing major programs in six years," noted Kaminski, not the 12 to 15 years that is so common today.
The veteran acquisition expert, who recently co-chaired a study on the importance of systems engineering in the early phases of defense programs for the National Research Council and is now chairman of a new study at the NRC on "Experimentation and Rapid Prototyping in Support of Counterterrorism, " was at pains to note that the defense industry suffers from many of the same problems as does the government. There are too few well trained systems engineering experts. There is too little prototyping and too little focus on subsystems.
Both groups would be helped by hiring more people with extensive systems engineering experience as well as spending money to train a new cadre of systems engineering experts. That will require hiring substantially more people for the acquisition corps, Kaminski said. He pointed out that cuts to the acquisition workforce began on his watch. He resisted it, but Congress was eager for some sort of peace dividend and would not let the issue alone. "When you reduce it by 40 percent you don't get rid of the people you want to keep. And we have not recovered from this," he told me.
In the next piece about my conversation with Kaminski, we'll look at what concrete policy should be made during the next administration, especially in connection with the broken requirements process. (Hopefully, I'll find a better picture for that one.)
