Don’t Swing Between Just-in-Time and Just-in-Case: A Better Framework
(This article originally appeared on the theEDGE1)
Whenever a natural disaster causes product shortages, you’ll hear people say that supply chains need to move from just-in-time to just-in-case.
The implication is that the supply chain was too lean and didn’t have enough inventory to meet the unexpectedly rapid increase in demand.
Coming out of 2020/21, some companies took this advice and built up their just-in-case inventory. Now, facing decreasing demand, I’ve seen a few stories about how the just-in-case inventory is now viewed as excess and obsolete inventory.
In other words, just-in-case inventory can quickly become a drag on financial performance.
Unfortunately, the terms just-in-time and just-in-case are catchy but don’t give us real guidance. Here are two misconceptions that lead us astray.
First, the just-in-time is now used to just mean keeping a bare minimum amount of inventory. But, the idea came out of Toyota’s overall lean manufacturing strategy. In the overall lean manufacturing approach, Toyota did hundreds of things that allowed them to run efficiently with low inventories. If firms just implement low inventories without understanding all the other aspects of the system, it will cause problems.
Second, when some suggest just-in-case inventory, it often comes with the hidden assumption that it would have been easy to predict what was needed.
There is a great line toward the end of Goldratt’s The Goal where they are talking about how to win a big deal. The sales leader says “...if we had only had the foresight to build a finished goods inventory of Model 12’s while we had those slow sales months…”
By this point in the book, the lead character is smiling to himself knowing that they have warehouses full of products that they’ll never sell. They never would have guessed to make this product.
Hopp and Spearman’s great textbook, Factory Physics suggests a better way to think about this problem in one of their laws: “variability in any production/operations system will be buffered by some combination of inventory, capacity, or time.”
In other words, you’ll buffer against variability by having inventory, having the capacity (to quickly recover when there is a problem), or with time (by making your customers wait for you to recover).
For example, if you don’t buffer against variability by having inventory or extra capacity, it will default to time (you just won’t be able to meet demand). You don’t get a free pass.
Also, the more variability you have, the larger your buffers need to be.
I think it is important to think about your buffer strategies for normal times vs natural disaster times.
In normal times, you experience variability because demand forecasts aren’t perfect, machines break down, shipments are late, and so on. In normal times, inventory is a fine buffer. You might keep a few weeks or a few months of supply, but it doesn’t create an unusual drag on your supply chain.
In this case, you don’t reduce inventory by carelessly implementing a just-in-time system. Instead, the observation above suggests that you work to reduce all the sources of variability and inventory can naturally settle at a lower point without impacting your business. And, by the way, this is what Toyota did over many years.
In natural disaster times, the variability is extremely high. That is, demand for certain products may go up by 10X, all the ports shut down, or factories, suppliers, and regions shut down.
It is in these times when I hear the most about just-in-case inventory. But, the variability is so high, that the amount of inventory that would have been needed could easily turn into a financial problem that threatens the existence of the company if the event doesn’t happen.
That is, if a company builds inventory for these situations, there is a good chance that the inventory will be for the wrong product, go obsolete, be damaged, spoil, or bankrupt the company before it is ever used. Clorox Wipes saw a sudden increase in demand of 500% in 2020. In 2010, if they had started to fill warehouses just in case demand increased by 5X, that would have been too big of a drag on the company.
Instead, the Factory Physics law offers another way to look at the situation. Since these variability events will happen, we still need to choose the buffer.
Toyota offers a good example. In 2016 the WSJ wrote an article after an earthquake shut down suppliers causing a production halt. Japan has been hit by a lot of earthquakes. You might think that Toyota would naturally adopt a just-in-case approach.
Nope.
Instead, Toyota was good to have a small buffer of time– they were willing to halt production. And, their big investment was in extra capacity. They said: “...we have been focused on how to understand and identify issues in case of a problem, and how quickly we could recover.”
Being able to quickly recover means having alternative sources of supply, having ways to shift production, and making an investment to understand critical suppliers deep in the supply chain. None of this is free, but they thought this was a better investment than excess inventory.
Every case will be different and requires hard work to create a strategy. There are arguments to be made for stockpiling some critical raw materials. But, companies shouldn’t shift to just-in-case inventory without thinking through all the ways to build in extra capacity to react to natural disasters.