In a recent post, I argued that DoD programs require long-term predictions about costs and requirements that get locked in as a “baseline.” But once a major program starts, it’s so well protected that just about anything can happen and it is unlikely to get canceled. So the cynic in me thinks: why bother pretending the baseline was real in the first place? This is especially true if waterfall/industrial planning processes are a major contributor to poor program performance.
Consider the following hack to weapons acquisition. DoD officials could knowingly create an acquisition strategy that they know isn’t real, but likely for approval. Just go through the motions, analyze historical costs and design to a slightly better version of existing systems with absolutely no intention of going that way. Line up the necessary information, get it approved. Now, break the entire development back down into partitioned contracts and incremental learning steps, and veer whichever way you want. It’s virtually too late to be stopped once funds obligations are ramped up.
This is the opposite approach of the Air Force’s ABMS program. Don’t overtly tell oversight agencies you want “10-15 percent solutions” and that the program will “emerge” and that it won’t be treated like a regular “program of record.” The hammer will hit you before you get anywhere at scale.
The dangerous cynic in me thinks: just say you’re doing JSTARS Recap. Get through your AoA, LCCE, Affordability, etc. Then, the first thing you work on is networking. All networking for risk reduction, and no aircraft bc that’s high TRL. Then, very slowly, it becomes apparent that there never was a manned aircraft involved. You offboard the capabilities onto other disaggregated or convenient platforms.
The great part is, you were fully funded to go experiment at scale! No fighting for years with the oversight complex. By the time they realize what has happened, the program is fielding all sorts of capabilities that the military loves and proves, through operational test and evaluation, that it is far superior to the legacy system. Boom!
There’s also a money and time advantage. If the Air Force went with JSTARS Recap, it had planned $2.4 billion through the FYDP between FY 2015 and FY 2019 (here and here). That ramp up may have been delayed by a year or two with AoAs and the like, but it would’ve been big figures by FY 2017 or so. That’s a huge advantage compared to what ABMS actually got — $27 million and $35 million provided in FY 2019 and FY 2020, asking for $302 million in FY 2021 but likely to get just about $250 million. Of course, there’s other sources of funds supporting ABMS, but still, it isn’t getting the funding to do the enormous task its given.
The funny thing is, all this is possible so long as no one realizes that its happening. Willfully deceiving the oversight complex will get you into hot water. Boiling water. And its infeasible to think that the “secret” will be kept close, since layers of bureaucracy are involved they’d all have to be “in” on it or misinformed. There are just too many reasons that this scheme provides undue risk to one’s personal interests. And so, the DoD is stuck with industrial era processes that forbid the kind of incremental learning that is necessary for real progress and force structure transformation.
Leave a Reply