And again, we have a crisis on our hands that raises questions about just-in-time or JIT. Currently, the measures to restrain the coronavirus have led to Chinese factories being closed. Consequently, supplies to other factories around the globe have come under pressure. Delays in supply could shutdown factories with massive financial consequences. But this is not a new situation of course. We have seen very similar discussions in the case of hurricanes, tsunamis, volcanic eruptions, earthquakes, nuclear disasters and, also recently, the BREXIT. So, is there a balance to be struck between just-in-time and resilience? Or can you build a resilient JIT (just-in-time) system?
Interdependent Supply Chains
Nowadays, both manufacturing and service industries rely on a complex supply chain, typically consisting of hundreds of suppliers in multiple tiers, and in many different countries. So, the dependency of organizations on other organizations has grown simultaneously. And a problem in any of these supply lines could spell trouble for you as a customer.
However, this is not only the case for a major crisis such as the current outbreak of the coronavirus, BREXIT, and the likes. This is just as much the case for a “simple” bankruptcy or a fire at a supplier location. The supply of materials is always at risk.
In fact, companies have always been and will always be at a certain level of risk. The question is not so much whether they are at risk, but how they go about managing these risks. Interestingly enough, during the 1950s, exactly Nissan and Toyota departed from the traditional strategy in automobile manufacturing: they increased their dependence on firms over which they had little or no financial control.
It is important to understand that Just-in-Time, or JIT, is a multi-faceted concept. At first glance, JIT shows itself as a system that aims to achieve a near-stockless supply chain. JIT ensures that materials arrive “just-in-time”, meaning not too late, and not too early either. It is often contrasted to a “just-in-case” system where products are bought and produced to cover for every case conceivable. Just-in-time systems are considered as systems with hardly any inventories, or buffers. This is even where Lean derives its name from, as it revealed itself as a near-bufferless system. And when organizations operate with virtually no buffer, it seemingly makes them more vulnerable.
However, JIT also is a process or production control system. JIT aims to synchronize demand and supply by supplying only goods that are actually required (e.g., by replacing materials when they have been actually consumed), instead of planned to be required by the (internal) customer. The well-known kanban system is the means to implement a JIT system, whereby the number of kanban controls the amount of materials that are present in the supply chain. JIT systems typically result in controlled inventory across the board, whereas traditional “just-in-case” push systems typically lead to uncontrolled situations with a lack of materials in one corner of the company and surplus of materials in another.
Finally, JIT is a system of continual improvement. As you can deduct from the above, the number of kanban controls the level of inventory in a JIT system. And to be able to reduce the number of kanban (and therefore the level of inventory), you need to eliminate the underlying reasons for having the inventory.
This last aspect shows that, although near-zero inventory levels are the ultimate aim of a JIT system, any JIT system should first and foremost be dimensioned at the levels of inventory required to maintain flow in the light of the problems it faces. The improvement perspective of kanban only tells us not to hide away from these underlying problems, but to face them, headfirst, and to eliminate these one-by-one, hence the name continual improvement or kaizen.
Sizing Your Safety
So, although JIT may ultimately aim to achieve a near-bufferless supply chain, it does not prescribe slashing inventories at all costs. Its policy, in fact, is to size these at the levels required, and to then progressively eliminate problems that allow for the controlled and safe reduction of inventory without penalizing customers.
In the JIT system, a distinction is typically made between safety and cushion stock. Safety stock can be seen as the money on your current account; cushion stock as your savings account. Safety should cover the “normal” variations in your system; inventory from your cushion stock can only be withdrawn in “special” cases. Cushion stock sometimes is even located in a different location than the supplier or factory location itself.
And how much inventory is enough anyway? Again, don’t only think about major crises like the current coronavirus outbreak, but also about more probable risks like bankruptcies, fires, or even major equipment breakdowns at suppliers. How much inventory to carry for each and every possible risk? And to carry this inventory at all times as you never know when Murphy will strike…
I’m not saying that you should not think about managing your risks. However, in doing so, many more factors come into play other than just buffering with inventories or adding redundancy. As these also increase the cost of doing business.
Sure, a JIT system that is near-bufferless is a vulnerable system. But it is so by design, intentionally. As it aims at progressively emerging problems by continually reducing its buffers. But inventories are very much in control. I even dare to state that traditionally run supply chains are at more risk than JIT-operated ones. Because JIT-ones have more control over their inventories. Traditionally operated companies are continuously at risk due to their production control system that leads to unbalanced stocks.
But not only this, JIT-operated organizations typically also possess many other characteristics that make them more resilient. JIT, or more generally Lean organizations, typically have both a highly reactive and thorough problem-solving culture. This means that Lean organizations, also by design, are able to significantly reduce the time that the disruption actually impacts the organization. Their ability to recover, and the speed with which they do so, is often impressive.
Even more, JIT-oriented organizations front-load their processes, i.e., they put a lot of attention not only in the design of their (standardized) components which makes them more flexible, but also their supply chains. They often opt for a single source strategy with carefully selected and competent suppliers for whom they are an important customer, and in which they sometimes even have a financial stake. They consider their suppliers true long-term partners, instead of the short-term, arms-length relation that often exists in more traditional supply chains. This creates a strong interdependency between customer and supplier, a strong loyalty towards the customer, and a high reactivity towards joint problems.
They also have a strong local supply base, instead of a globally dispersed supply base. This makes them only vulnerable in a specific region, and not globally.
In conclusion, it can even be said that JIT is in the news again now, exactly because JIT reveals problems. It was designed to do so. But we typically misinterpret JIT, as a system that, in whatever circumstance, hardly has any buffers. That is not true. It has the buffers it requires given its risk level. At the same time, however, true JIT organizations possess far more characteristics than just inventory buffers that make them more resilient than traditionally operated organizations. The fact that an organization is Lean, therefore, does not necessarily makes it brittle. On the contrary.
again a nice article introducing some stuff I didn’t know yet.
About the cushion stock you explain – I reckon it is also set up with a kanban system to avoid ageing of the stock?
Thanks for your comment. Good question about the cushion stock. In one of my previous roles, I have been confronted with cushion stock (or strategic stock as our customer called it). It was located off-site and to be honest, it was “a pain in the ***.” It is very difficult to maintain FIFO in such a setup, and risks are continuously present due to this (think of design changes, defects found at a client site, or proposed product improvements). You can either set it up with kanban as an additional node, but we flushed it at a certain interval. In any case, you are confronted with increased cost and risks ALL the time. And we only had the required three days of this cushion stock anyway. So how would that have helped us in a serious crisis?
Another great article!
I agree that JIT is a system to synchronize the company and to bring out the problems: it is by gradually reducing the variability of the system that we can approach the ideal of “zero stock”. In the meantime, it is necessary to buffer the variability with the 3 classic buffers (see factory physics): stock, capacity, lead time.