This characterization of the roots of PM [project management] represents a certain irony: the Manhattan Project did not even remotely correspond to the “standard practice” associated with PM today. Indeed, the Manhattan and the first ballistic missile projects fundamentally violated the phased project life cycle approach. Both applied a combination of trial-and-error and parallel trials in order to “push the envelope”, that is, to achieve outcomes considered impossible at the outset.
However, the project management discipline has now so deeply committed itself to a control-oriented phased approach that the thought of using trial-and-error puts professional managers ill at ease. In our seminars, experienced project managers react with distaste to the violation of sound principles of phased control when they are told the real story of the Manhattan Project (or other ambitious and uncertain projects). The discipline seems to have lost its roots of enabling “push the envelope” initiatives, de facto focusing on controllable run-of-the-mill projects instead.
That was from an excellent paper by Lenfle and Loch”Lost Roots: How Project Management Came to Emphasize Control Over Flexibility and Novelty.” The US pursued parallel paths. For fissionable material, there were two plutonium and three uranium paths concurrently taken. For the bomb design, there were four paths. But it isn’t just parallel paths that matter — it is keeping options open to modify plans, start new paths, and cancel others. Here’s Lenfle and Loch again:
For the production of fissionable materials, a breakthrough came when it was discovered that a new process, thermal diffusion, could provide slightly enriched uranium, which would then feed the gaseous diffusion and electromagnetic processes for further enrichment. The parallel processes were unexpectedly combined into a composite process that finally achieved the desired performance.
For bomb design, a second group of scientists had worked on an implosion design as a back up. When it became clear in the spring of 1944 that the gun approach did not work for plutonium, the implosion design became first priority.
Think about all the talk on rationalizing so-called “redundant” or “duplicative” projects in the services like in AI/ML, JADC2 and enterprise tooling, hypersonic vehicles and long-range fires. These duplications are only ineffective if there is relatively complete knowledge about how to best achieve them. Because the DoD is still in the early days of experimentation, these duplications are really a discovery procedures about what works and doesn’t work, as well as an rivalrous incentive to get there first. Moreover, each team approaching the problem differently may learn from the other team, greatly accelerating the advancement of the whole.
Leave a Reply