Almost 3 years ago.
Focusing Steps
Also don’t let the non-constraints produce more than the constraint can process.
If you do let that happen then you’ll build up some useless inventory that might be clogging up space. Maybe you’ll have to throw away something you spent time, energy and capital on producing.
Excess Capacity or Balanced Plants
- variance and statistical fluctuations
- a balanced plant aims at 100% utilization and 0% idle time
- (simplified) every workstation would output the same amount
- some stations would purposely do less than they could to fit in
- (simplified) every workstation would output the same amount
- assembly lines means there are dependencies, i.e. station C gets its parts from B and B gets its parts from A
- A produces more than B: we get extra parts that clutter things up
- A produces less than B: B has wasted potential, it can only produce as much as A gives it. A is the weakest link and the chain can’t get stronger if we don’t strengthen A
- to avoid this we can create a buffer in front of B such that it can work when A doesn’t produce enough
- we can also increase the production capacity of A but stop it when it produced everything that B needs
- buffers help some stations keep up when the stations they depend on are underproducing
- bigger buffers give more safety but take up more resources so it’s a tradeoff
- feeding stations have to halt work when the buffer is full
- some stations can have excess capacity such that it will rarely produce less than what is needed for the next station
- it will have to limit its output most of the time in order to not build up excessive buffers everywhere
- this is an unbalanced plant. some stations have more capacity than others
- relying solely on excess capacity and not using buffers carries risk because A can have really bad day or be completely broken
- with more excess capacity you need less buffer and vice versa
- there’s a tradeoff and we mostly decide based on how expensive each are
- with more excess capacity you need less buffer and vice versa
- a balanced plant which has a bad day will on average not replenish its buffer
- A should have more capacity than B uses. it doesn’t have to be a lot more. the buffer can be gradually replenished if the average rate of A is higher than
- buffers deal with variance and excess capacity replenish the buffer
- if C has bad luck extra inventory can pile up ahead of it
- in order to reliably get rid of this inventory C must have excess capacity in order to get rid of it
- the amount of excess capacity doesn’t matter much, C will catch up pretty quickly anyways
- everything has variance
- you can reduce variance but not get to 0
- design the plant around the bottleneck
- there’s always a bottleneck so don’t try to avoid it, don’t try to build a balanced plant
- multiple stations tied for least capacity only makes the system chaotic due to variance
- the bottleneck should usually be the most expensive thing and since we need lots of excess capacity they should be cheap
Which workstations need buffers? Only the bottleneck.
why?
Bottlenecks are the ones that hold the rest back. If some station early on in the line underproduces then it can cause all the rest of the stations to be underutilized. But when it reaches the bottleneck it can just start using from its buffer and from there on normal production can continue.
The buffer is there to make up for other stations underproducing.
And which workstations need excess capacity? Every non-bottleneck.
because if one station in the chain has less or same capacity then that station can become the bottleneck. the stations after this station can’t produce more than it produces. this could either cause the original bottleneck to stagnate on its buffer size or use up the buffer.
A plant may have multiple bottlenecks if it has multiple production lines. It can also have a more complicated production line that isn’t a linear chain. E.g. there could be three workstations that make parts which feed into B, then B combines all those incoming parts, and then the output of B feeds into multiple later workstations. There could also be multiple B workstations, and they could have different production characteristics (e.g. one uses a fancy new machine, another uses an older machine, and a third uses hand tools). These details complicate the analysis, but the principles and conclusions remain similar.
- analyze more complicated scenarios like these
This is related to decision making in general. Usually there are lots of factors that are easy to get plenty of, and only zero, one or a few factors which are hard to get enough of. With factories or life in general, if there are many factors which are hard to get enough of, what you’re doing may be too difficult, and you should give some consideration to changing approaches.
- Think of things this can apply to
related thoughts
The most important thing to optimize in life is what to optimize. I.e., to optimize your focus. Focus on what matters. Keep in mind what your ultimate goals in life are. Don’t spend too much resources on minor goals or sub-goals.
Having a bunch of money is a bad ultimate goal in life. Money doesn’t have intrinsic value. If you don’t use it for anything then it’s no good. Having money is a sub-goal for other ends. But making a bunch money, the act of producing, could be a great goal.