A Short Supply of Long Thinking – Part 1
Category: Rebuilding the Supply Chain • Oct 21, 2020
Some in the media blame the supply chain disruptions during the pandemic on just-in-time production or JIT (e.g., “Why Are There Still Not Enough Paper Towels? Blame Lean Manufacturing”, Wall Street Journal, August 21, 2020.) So IMTS reached out to a champion of the Toyota production method.
Dr. James P. Womack, Ph.D. is the father of the lean movement and founder of the Lean Enterprise Institute. His seminal 1990 work The Machine That Changed the World made lean production known worldwide . Dr. Womack—“just Jim” to his friends—sheds light on the true causes of our current state shortfalls, explains why they were inevitable and suggests some much needed remedies.
IMTS: What was the state of the U.S. supply chain before COVID-19?
Jim: If you go back to OEMs such as GM and Ford in their glory days, they were vertically integrated with an internal schedule for parts supply. The supply chain was an internal thing, rather than external.
By the 1980s, as companies felt global pressure, there was a general sense you could have more leverage negotiating with an outside supplier than you could with an inside supplier. As a result, big companies started to de-integrate. When they shifted to an external supply chain, they lost control over many factors and then needed to rigorously manage the flow of materials and information. At the same time, OEMs embraced materials requirement planning (MRP) and attempted to perfect it, to the extent that MRP can ever be perfected.
And CEO’s everywhere were feeling competitive pressure and looking to inventory reductions as a way to reduce carrying costs and generate cash. For example, I visited Home Depot in the early 2000s during the reign of CEO Bob Nardelli. He had decreed that Home Depot would take down inventories 20% at constant sales without changing anything about his supply stream. The result was massive out-of-stocks on the shelves. So, Home Depot called me and Dan Jones to ask how Toyota would take out inventories to free up cash. The reality we found was definitely mean but not at all lean.
IMTS: How does MRP compare with the alternative, JIT?
Jim: MRP is a push system. Think of it as a central computer that figures out what ought to be sent where and when along complex value streams involving many facilities and companies. All the information on the current state flows goes to the central processor and then instructions on what everybody ought to do next flows out of the processor.
Toyota consciously set out to introduce a pull system. The rule was that there would be a single order point near the customer where demand would be leveled and that downstream processes would order from upstream processes exactly the amount that each downstream process had just consumed, and this became known as JIT.
You can run a successful business with either JIT or MRP, but both take discipline. JIT requires discipline with regards to demand leveling at the order point and carefully controlling the number of instruction cards in the system. MRP requires discipline with feedback loops to make sure you get real information, as opposed to optimistic information or gaming of the system by managers trying to make their numbers by ordering just-in-case inventories.
IMTS: The world thought that inventories would be lower just by applying JIT, didn’t they?
Jim: That was a complete misunderstanding. Toyota calculates what it calls standard inventory for every step in a complex process. This consists of safety stock at every step to keep from starving the downstream customer, shipping stock for the amount of inventory being built for the next shipment and buffer stock to protect each step from erratic supply by the next up-stream process. And the standard inventory is what it is, based on data about recent operational performance. There is no way to reduce inventories except by improving operational performance. Cuts like those at Home Depot are madness.
You would think the tendency of a JIT system would be for the total amount of inventory to get smaller over time through operational improvements. But in their absence, the amount of inventory is not going to be very different from an MRP system.
IMTS: What happened with COVID-19?
Jim: There is a fourth category of inventory in the Toyota world called emergency stock that is outside of the JIT system. That system is only designed to deal with “normal” variations in demand (e.g., more cola sales on a hot day, fewer sales of cars during storms in the winter). Emergency stocks are for demand surges that can’t be anticipated – tsunamis, earthquakes, pandemics - and are typically finished goods stocked close to the customer. In the event of a tsunami or a pandemic, the provider has a pile of ready-to-go inventory (or reserve production capacity) proportional to the size of the wave.
This is challenging because such events are hard to predict. It’s fiscally impossible for most private entities to support a large emergency stock and really hard for politicians to manage, so companies and nations tend to have too little emergency stock. Emergency stocks also need to be secure and well-managed with regard to allocation rules. Otherwise, it just becomes a free-for-all, and that’s what we saw most recently.
IMTS: Some in the business press blame JIT for the current supply chain problems.
Jim: That is just nonsense. Recent supply chain disruptions are due to enormous spikes in demand and not having the appropriate emergency stocks in place. They have nothing to do with either JIT or MRP.
And even if JIT was a problem, there’s actually not very much JIT in the rigorous Toyota sense. It is still mostly an MRP world. Other than Toyota and its network of suppliers, disciplined JIT is notable for its absence, and yet the media accounts always presume that the world switched to JIT. The switch didn’t happen.
IMTS: In light of COVID-19 disruptions, what should the manufacturing world focus on?
Jim: Every industry has a collective interest in not getting themselves into these messes. Wouldn’t you think that all the people active in the same field should collaborate? Someone needs to say, “Whoa, we don’t want to get our tail caught in the door when things go south again.”
By the way, allocation within nations is pretty chaotic, but between nations, it is the Wild West. A supply chain may have the capability to substantially increase output through emergency stocks or reserve capacity, but in an international supply chain, each step is subject to being diverted by another country.
IMTS: Suppose a country concludes it can’t control an international supply chain. Then what?
Jim: That indicates the country needs a larger emergency stock. Unfortunately, we always have this discussion after the fact rather than before. I wish everybody would put their heads together—and we could do it country-wide and by industry—and agree on what the strategically critical items are (e.g., medicines, PPE, test kits, ventilators in the case of a pandemic) and how to create and allocate an emergency stock. I guarantee you this is not the last time we’re going to see a disruption of this magnitude.
IMTS: Why does it seem we are always underprepared?
Jim: Because we don’t really know how to calculate the odds, and then we don’t like to think about it. How many emergency reserve supplies do countries on the Pacific Rim have for the next tsunami? They need to make those decisions now. They might be right or wrong, but any decision is better than no decision, which seems to have been the system in much of the U.S. with regard to medical equipment.
IMTS: Are we more risk-prone today?
Jim: I don’t think the world is a more fraught place than it was in 1990 or 1960. These black swan events have always been there because the world is a complicated place. From time to time, the awareness of how vulnerable we are gets amplified, but then we slowly minimize awareness of our vulnerability. It takes a conscious effort to discuss how we need to support important supply chains. What’s the plan? Who’s going to do what? Who’s going to pay for it—which is the real issue. How optimistic am I about the U.S. preparing adequately in the future? Not so much.
IMTS: What about willingness to execute a national manufacturing strategy?
Jim: Well, the historical record has not been good. There’s a thread in the national conversation that if 100% of the supply chain had been within our borders, we would not have had a problem with PPE and ventilators. Not true. First, we had insufficient emergency stocks, and second, we had insufficient reserve manufacturing capacity because nobody really thought through a disaster scenario.
I worry about naïve thinking, such as saying that if ventilators have 90% domestic content, then we won’t have any more shortages. Logically speaking, those two statements have nothing to do with each other.
IMTS: The problem is more complicated than people realize. All it takes is one missing control board, and a machine becomes an inert object.
Jim: Absolutely. It requires a detailed discussion about how we will have availability for every component of a complex product.
By the way, there’s always an alternative for uncertainty in a complex supply chain: have an emergency stock of ready-to-go units. My general bias is toward the emergency stock of finished units with a very clear allocation scheme.
IMTS: Should we use different strategies for short-term vs. long-term needs?
Jim: Needs that are temporary spikes, such as ventilators and masks, you can meet with emergency stock and a clear allocation system. That’s different from a military scenario where you need to prepare for unanticipated and lengthy conflicts. In that case, you need the capability for emergency, end-to-end output increases, and you need to have production in safe locations in order to achieve it.
For part 2 of this interview with Dr. Womack, click here.