Basically research and procurement are incompatible. New developments are upsetting to procurement standards and procurement schedules. A procurement group is under the constant urge to regularize and standardize, particularly when funds are limited. Its primary function is to produce a sufficient supply of standard weapons for field use. Procurement units are judged, therefore, by production standards.
Research, however, is the exploration of the unknown. It is speculative. It cannot be standardized. It succeeds, moreover, in virtually direct proportion to its freedom from performance controls, production pressures and traditional approaches.
That was Vannevar Bush testifying to the House Military Affairs Committee after WWII, found in Converse’s excellent Acquisition History Vol. I. Here’s the other side of the debate:
On the other hand, Lt. Gen. W. H. Simpson, president of the board of officers that made recommendations leading to the June 1946 reorganization, disagreed. “Procurement and research,” he wrote Eisenhower in February, “must go hand in hand if the latter is to pay dividends.” He further stated that the board he chaired “does not concur in the idea that [research] has no connection with production.” Although advocates of separation recognized that a relationship must be maintained between the two, many, like Simpson, believed that any geographic or organizational separation would hinder innovation and slow development.
My inclination is to agree with Vannevar Bush that R&D and procurement should be separated, and that the former seeks to disrupt the latter. But then again getting R&D into procurement to accelerate iterative learning in an agile/lean methodology way is also important — as Elon Musk has argued for Starlink production.
I think the software- and data-intensive nature of current systems is pushing me even more clearly onto the need to combine more of R&D and procurement. In DoD, the way this is done is through a “program element” construct which has both RDT&E and procurement appropriations. But this leads to arbitrary phasing and a need for program offices to get immature systems into full-rate production just because the money is looming. Then the separation real separation comes between S&T in the labs and development in the program offices.
I think one of the challenges in acquisition is understanding this tradeoff between separating or combining R&D and procurement as software practices like agile and devops start becoming more important to systems overall.
Leave a Reply