2013-09-19 After his presentation to the leadership of the 33rd Fighter Wing, Secretary Wynne was asked to discuss the challenge of what is called concurrency.
The answer by Wynne provided one of the best understandings of the reality of the approach taken in the modern aerospace industry.
There would have concurrency not matter when you started the process.
Because we tend, in industry, to hire to a very tight line; enough to get the job done; but not enough to be accused of introducing an overrun. As a result we do not see the funding to the full up line that industry would like to maximize efficiencies.
And industry will not put people on until they get slightly behind schedule. This is because we’re so worried about people cutting the program back, as the start is usually contentious, and up the line customers threaten to leave us high and dry and having to lay off a bunch of recently hired people.
And the other thing is one does not discover many problems until we get later in the program. The top-level design is roughly perfect; with the devil in the details. This is called integration.
Concurrency tends to sway like a pendulum of a clock from we want to involve the users early because we want the user feedback, and we want the engineers to get beat up and understand that they screwed up in the design. This is called direct feedback.
But you can’t get that if you wait, wait, wait, wait, and then have the tests and all your engineers have gone onto other projects, and they never actually meet the user because we waited so long.
And then the other side of it is, if you waited, would you really have solved that problem?
I don’t know.
It is a question of balance. Every program manager is going to be subject to demands to meet the IOC as quickly as possible versus counter demands that they should’ve waited and fed in changes to airplanes number one through twenty before going operational. Only when top leadership takes overt possession of the Program Manager’s (PM’s) dilemma is it called concurrency.
We will always want to feed in the air changes to airplanes one through twenty.
But doing development without deployment guarantees you will not have a new asset out there reshaping capability.
It also guarantees that the impact on operations will be shaped by testers, and not by operators.
For earlier looks at how to understand concurrency see the following: