The idea that growth can come from a focus on limits, rather than limitless expansion, David Epstein explains (in Inside the Box), is at the core of a simple but profound idea:
In the 1970s, Israeli physicist Eli Goldratt was working on algorithms to predict the behavior of atoms in heated crystals, when a friend asked for help with a comparatively quotidian problem: constructing chicken coops.
[…]
Goldratt realized that shifting a lone extra worker from a non-bottleneck over to the bottleneck could multiply the total output of the business. He crafted a new schedule that actually called for some workers at non-bottleneck steps to work less. If they worked at full capacity, they just produced more parts that piled up wastefully before the bottleneck. Without changing the workforce, the new schedule tripled chicken-coop output. This was the beginning of what Goldratt would call the “theory of constraints,” or TOC: the notion that a single bottleneck determines the fate of a system.
[…]
In 2011, Time magazine named [The Goal] one of the twenty-five most influential business books, alongside genre classics like Dale Carnegie’s 1936 How to Win Friends and Influence People. In 2013, Jeff Bezos hosted an all-day book club on The Goal with Amazon executives.
[…]
But for managers judged only by the efficiency of their particular stage in production, there was little incentive to change. Goldratt had a saying: “The sum of local optimums is not equal to the optimum of the whole.” He saw that separate parts of an organization had their own incentive structures, all set up to optimize production in their silo—like Boy Scouts running ahead of Herbie—even when it was wasteful for the whole.
Long-time readers might know that I’ve discussed the Theory of Constraints plenty here, just not recently. Kevin Fox‘s Blue Light anecdote is an excellent soft introduction to the key concept. I discuss a more detailed example from The Haystack Syndrome in another post, if you’re interested in a bit of a puzzle, with numbers.
Epstein offers a few unconventional examples that don’t quite match Goldratt’s concept, including this one:
In one theory-of-constraints case study I encountered, a factory that made custom gearboxes for industry had a severe bottleneck, not on the factory floor, but in the fifteen-person design office where the gearbox plans were conceived. The designers had so many projects in progress that their workday was ravaged by multitasking. These were intricate endeavors, and the designers were switching focus more than fifty times a day. It led to errors, and then to quitting. Ultimately, the design team implemented a rule: “Stop starting and start finishing.” Orders kept coming in, but the design office was forbidden from working on any new order until one already in progress was finished. Multitasking plummeted, and a few months later the office was getting three times as many designs out the door.
This is more an example of Goldratt’s concept of the Critical Chain, or classic project management.
I’m a simple man. I see Goldratt, I feel compelled to wade into the comments section, even if I have little to add that the author has not already said.
Overproductive workers who produce subassemblies are an example of physical constraints of part storage. Real-world factories don’t have infinite buffer space to store subassemblies. Overproduction is a problem for many reasons, but if we had some kind of Star-Trek-tier space warp for infinite storage, overproduction would be less daunting. But even if we had infinite buffers, Murphy’s Law would guarantee that some businesses would still find ways to go bankrupt.
Regarding engineers: “This is more an example of Goldratt’s concept of the Critical Chain, or classic project management.”
Maybe that is right, but I would need quite a lot of time to dig into the details and mentally digest the case study.
My hot take, prior to digging deep, is as follows: The problem with engineers who get too creative is more akin to managing painters or sculptors or musicians. Engineers are very creative, and creative people don’t maximize profits by exploring creativity. The “buffer space” for creative engineering ideas is so large it seems to be effectively infinite. Obviously, intensive exploration of creative ideas is not the same as overproduction of physical subassemblies.
I suppose I should have linked to my fourth Critical Chain post, which addresses multi-tasking.