[peteradio]

I worked at a place with a lean 6 sigma certified specialist who towards the end of the companies doom effectively had the lead engineer cleaning out molding machines to track down every last tiny molded part that over the course of several years of continuous running had flung outside of its target. Same guy told me if the coke machine ever stole my change that he'd help me get it back from the vendor.

[7thaccount]

All the lean sigma stuff seems like another useless management fad to me that only benefits consultants. Is that what you're saying here?

[thereddaikon]

Like many management system fads, it started as a useful kernel of wisdom or obvious maxim that idiots ran with and turned into a monster. In the case of six sigma, the idea is about constantly optimizing your workflows and not accepting "we've always done it this way" as an excuse.

But most people lack the critical thinking skills to correctly apply wisdom when necessary and instead need a solid framework to operate within. That's how these things inevitably develop. Just like how Agile is supposed to be about getting working code over being bogged down in process but inevitably ends up with with half baked products that have massive issues.

Six Sigma also gets applied to industries it has no business being in. The mentality works best when you have a fixed workflow. In manufacturing it would be if you are making a lot of one thing. You can do a lot of optimizing. But I have seen it employed in organizations where every project was vastly different. Instead of a lean approach it should have, and previously was, following a house of quality philosophy. This happened to be in an industry where cost was rarely a consideration but performance and reliability were.

[7thaccount]

Great comment and thank you. It has only popped up sparingly in my industry thankfully.

What you're saying makes sense to me in that you should never fall into the "this is how I've always done it" trap, but putting a complex beauracratic process around that is just going to create a whole new problem.

[citrin_ru]

> we've always done it this way

What was done for some time is at least to some extent proven o work. Any new way is not yet proven to work and can be worse (or can be better but it is not granted)

[Spooky23]

It’s an expression of distrust.

If you ever work in government, procurement people think like that because their goal is objective competitive process that that meets the minimum standard to fulfill the purpose.

There’s a certain logic to it. You don’t want to see random government employees driving around in Teslas, so generally speaking they will be in nondescript 4-door sedans. Having a human say “no” makes them accountable, so a complex process will determine what kind of car you need.

Taken to extreme, it becomes a problem. Procurement officers get lazy and focus on their process instead of the needs of the customer. So they treat humans like Ford Tauruses and allow vendors who understand how to game the process walk out the door with millions.

[randcraw]

From what I've seen after the attempt to adopt Six Sigma within a large pharmaceutical, 6S has several inescapable vulnerabilities due to not knowing:

1) which important company components produce signals that aren't measurable (e.g. internalities like deep domain knowledge in R&D or externalities like market swings or competitor behaviors),

2) which signals are too dynamic or complex to optimize, and

3) which signals are insignificant, irrelevant, irreproducible, or simply meaningless noise.

Applying Six Sigma to R&D illustrates these caveats nicely.

[peteradio]

I'm only speaking towards this one particularly useless buffoon, but the fact that he was allowed to wield any sort of power over anyone says something.

[doctor_eval]

This is the kind of thinking that is wrecking a business I’ve just quit. Management consultant decides that we need more customers so instructs us to open the product up globally. We explain that this is likely to be both expensive (lots of testing needed due to the usual geographic differences), but more important, based on our collected data and the use case (geographically local content), this expensive change is statistically likely to result in only a single new customer (literally, one).

But “every customer counts” so other activities which are likely to result in more customers are put on hold to capture a single customer.

The problem in these cases is often that the management consultant doesn’t consider the resource limitations and opportunity costs of their decisions, particularly if they come from a much larger business. If you have only one lead engineer, then getting them to chase discarded parts (or, in my case, a single customer) makes no economic sense at all.

Quite often, the problem is not a focus on measurement per se, but rather the very human problem of focusing only on those metrics that support the analyst’s intuition.