Search This Blog

Sunday, September 27, 2020

Comments on Goldratt video _Basics of Theory of Constraints_

In this video Goldratt starts out talking about that he was asked to write the intro to a TOC manual. He wanted to write one sentence that could be used to derive all of TOC. Then he decided to make it one word -- FOCUS.

Goldratt says that (my paraphrase): 

  • Regarding the 80/20 rule (or Pareto principle)...
    • The point of the principle is this: to illustrate that not all the things you could put effort in will provide equal amount of results.
    • The rule only applies if the points in the system are not interdependent. And Pareto said this but people ignored him. For an interdependent system, the situation gets more extreme to the point that only 1 thing is responsible for all the profit. (probably butchering what he said)
    • my note: It's super common for people to adopt a principle without understanding its scope (or more generally that principles have scope), and then misapply that principle to situations outside its intended scope.
  • Regarding the "prevailing notion" that when you see something beneficial, doing more of it is better:
    • More is better, only if it's a bottleneck. For non-bottlenecks, more is better up until a threshold, and then more is worse after that threshold.
      • The threshold is determined by the interdependencies between the bottlenecks and non-bottlenecks.
        • So you can't figure out the threshold when you are looking at a non-bottleneck in isolation.
          • Therefore, for the vast majority of a system, which are the non-bottlenecks, local optima =/= to the global optima.
  • Focus - do what should be done AND don't do what should not be done.
  • Cost accounting (at least the way business people understand it) incentivizes people to focus on non-bottlenecks.
    • The original article explaining cost accounting explained it's scope and warned against what conditions would make it break (not be useful for it's purpose of generating more profit). 
      • my note: that god damn ignoring-scope mistake again. 
        • looks like a cargo-culting mistake too. like somebody invents a great new idea and starts applying it. others copy it without really understanding how it works, and so they copy it all wrong.
  • People misinterpreted his idea of "process of ongoing improvement" and he contributed to the misinterpretation. 
    • He showed a graph of performance over time and 2 curves on that graph. Both curves are in positive growth, but the red is accelerating and the green curve is decelerating. Eli says most people think the green curve is more realistic. That's ridiculous. The history of economic activity is an accelerating curve, not a decelerating one.
      • Eli asked presidents of companies this question and many of them answered "green is the more realistic" despite their own companies doing the red curve.
    • The green curve is stagnation.
    • The financial performance (or growth performance) should follow the red curve.
    • The stability performance should follow the green curve.
      • part of stability is the creation of buy-in by the people involved. 
    • There's no end to improving stability. 
  • He recognized a problem, which is that TOC produced such vast knowledge, even within one company that implemented it, that it was too hard for people to understand without many misinterpretations resulting in no longer seeing the expected results of TOC implementation.
    • He wanted a way to organize the body of knowledge such that transferring the knowledge to a new person would be successful.
    • FUCK! I thought he was going to say what his solution was but the video ended without it.

No comments:

Post a Comment