Read on Twitter
Defense tech THREAD:
Everyone knows that the defense acquisition system has problems. One problem is that it’s overly complex has to work for everything from notebooks to F-35’s. That’s a tough optimization problem that requires trade-offs. (1/)
Everyone knows that the defense acquisition system has problems. One problem is that it’s overly complex has to work for everything from notebooks to F-35’s. That’s a tough optimization problem that requires trade-offs. (1/)
Therefore, the best way to reform the defense acquisition system is to focus on a narrowly defined problem and to implement corrections to address just that one problem. Problems and solutions vary based on what DOD is trying to buy. (2/)
Recently, though, the reform community has pushed for speed. Everything—from notebooks to F-35’s—needs to go faster! But speed means different things for different systems. (3/)
Recent congressional reforms (which broke up AT&L) pushed for speed everywhere, including on DOD’s most costly and complex major defense acquisition programs (MDAPs). (4/)
To increase MDAP speed, reformers reduced and decentralized the Office of the Secretary of Defense’s (OSD) oversight of MDAPs. On its face this makes sense—less checkers and more doers allows programs to go faster. (5/)
But OSD oversight also helps programs set realistic requirements and estimate schedules (and costs!) more accurately. From this perspective, OSD oversight might actually help DOD field MDAPs faster. (6/)
As it turns out, we can see this in historical MDAP data. Since 1962, MDAPs that were initiated during periods of strong, centralized OSD oversight were fielded faster and with less schedule growth. (7/)
This means that if DOD wants to increase MDAP acquisition speed, it should strengthen OSD oversight. Unfortunately, DOD is doing the opposite right now (hint hint…) (8/)
Further, since 1962, DOD has fielded MDAPs at speeds that are consistent with external benchmarks. This means that even though the entire acquisition system *feels* slow, at least for MDAPs, DOD is doing ok. (9/)
It also means that in the push for speed, reformers shouldn’t focus on MDAPs. In all but a few worst cases (which admittedly were very bad), DOD is doing fine. And DOD does even better when OSD exercises strong, centralized control over MDAPs. (10/)
Going forward, reformers should still focus on speed, but they should narrowly define their problem, collect data to assess that problem, and then develop targeted solutions. When we collected that kind of data for MDAP schedules, we were surprised by what we learned. (11/)
This brings me to my final point. Read my paper!
(cc: @CSIS, @CSIS_ISP)
(/END) https://www.csis.org/analysis/cycle-times-and-cycles-acquisition-reform
(cc: @CSIS, @CSIS_ISP)
(/END) https://www.csis.org/analysis/cycle-times-and-cycles-acquisition-reform
And for even more on acquisition reform, see Peter Levine's work. For example: https://warontherocks.com/2019/07/ten-rules-for-defense-management-reform/
