But how we build aircraft, that doesn’t make sense. Approaching NGAD [next-generation air dominance] the way we did the F-35 would put us at great risk. It would shrink the industry base even further and incentivize companies to get out of the fighter-building business.
The idea of the ‘Digital Century Series’ is not about building aircraft that are different, but about building aircraft differently. The key tenet is a new ‘holy trinity’ of technologies that would flip the pace of building new things and the price we pay for them. That trinity is: agile software development—no surprises, there—modular, open-systems architecture—because we want to be able to change out components quickly and seamlessly—and, finally, digital engineering, which is the new element.
That Air Force Acquisition Executive Will Roper, “Q&A: A New Way to Build Fighters.” HT: Chad Millette. I would argue that there isn’t much fundamentally new about digital engineering. Indeed, one of the reasons that prototyping was scaled back in the 1970s after DepSecDef David Packard was because computer-aided design started coming around. There was less need for trying things our empirically. Before that in the 1950s, it was believed that the same optimization principles applied but usually they were performed on slide-rules or punch cards.
It seems that as our engineering tools become more advanced, the environments which we simulate also become more complex. Even though digital engineering is without a doubt important, it does not replace the need for trying different concepts out in hardware.
Back to the agile/modularity points, Roper later says “You could build airplanes LEGO-style.” I’m sympathetic to this view, but it seems to follow that DOD would need to fund an ecosystem of component technologies from which platforms can rapidly integrate around. This reverses 70 years of acquisition policy around the “weapon system process” which takes the exact opposite view of technology progress, where advances come from tightly integrated designs and “single-shot” programs. It reminds me of JL Atwood’s testimony to Congress in 1959:
As defense technology continued to advance in succeeding years, it has become increasingly necessary to have early integration of each piece of equipment with all related equipment and the entire system. Therefore, it became desirable to develop groups of items together as integrated subsystems, which together make up what we call a weapon system. This trend was recently described by Ralph Cordiner, chairman of the board of General Electric Co., in these words:
“Where the need was once for a large number of general-purpose components and subsystems, the demand is increasingly for complete systems and even supersystems. The need for components of very high reliability and advanced design remains, but they must more and more be planned in context with the concept and design of the system of which they are to be a part.”
Thus there emerged the present pattern of weapon system development contracting.
If you take the digital engineering concept too far, then I think you run into the weapon systems concept that has dominated since the 1950s/60s. It presumes you can find the optimal specification in advance of prototyping, and it perhaps becomes the leading consideration for project selection. But when you balance it within a broader ecosystem of modular designs and agile processes, then digital engineering can be a huge benefit to the ones responsible for development.
Leave a Reply