On Friday, Feb. 28, 2020, technology executive Pierre Haren and I published a short article in which we predicted “that the peak of the impact of COVID-19 on global supply chains will occur in mid-March, forcing thousands of companies to throttle down or temporarily shut assembly and manufacturing plants in the U.S. and Europe.” This prediction was accurate; see, for example, the closing of automotive factories in Europe and the U.S. or luxury brands shutting down their manufacturing activities in Europe.
We are now six weeks after the publication of our article, and a few things are abundantly clear. As manufacturing activities in the European Union (EU) and North America are going offline, China is slowly recovering, yet at diminished capacity. At the same time, the pandemic strongly affects demand, not only worldwide supply. In the automotive industry, for instance, experts predict that the epidemic could mean “millions of fewer vehicles sold this year than earlier projections.” In the short term, the same report estimated a 17% to 18% decline in vehicle sales in March in various U.S. cities.
Research Updates From MIT SMR
Get weekly updates on how global companies are managing in a changing world.
Please enter a valid email address
Thank you for signing up
Most companies are trying to find their way through these tough supply and demand challenges. In parallel, they need to prepare for the recovery that will surely arrive in the not-too-distant future. Unfortunately, supply chain systems and processes employed by companies are designed for normal operations, not for a once-in-a-lifetime disruption and recovery. So the question is: How can a company fashion a recovery plan given the huge uncertainties that now exist on both the demand and supply sides?
Given the level of uncertainty, many scenarios are possible, yet it is instructive to focus on three key ones in order to develop some practical insights into the range of outcomes. These scenarios represent three different realizations of the recovery length and magnitude: the pessimistic, worst-case scenario; the more optimistic, best-case scenario; and the most likely scenario.
For the worst-case scenario, assume that no vaccine or cure is available for a long time, there’s a prolonged need to maintain social distancing measures, and a significant impact on demand and on the productivity of the manufacturing and transportation sectors occurs. In such a scenario, we are likely to see a growing number of bankruptcies of essential air and freight transportation companies as well as many small and midsize suppliers and manufacturers. All of these developments will make the return to normal a very painful process for consumers, investors, and manufacturers.
As for the best-case scenario, North America and the EU are able to control and reduce the pandemic through testing and social distancing, and life is back to normal by the end of the second quarter. In this case, expect significant pressure on logistics capacity, from transportation to warehousing. As companies from different industries will be scrambling to ramp up activities, a most probable bottleneck will be logistics capacity, which will be at a premium during that time. Today, logistics capacity shortages can be somewhat reduced as airline carriers use empty passenger jets to carry freight, but when recovery is in full swing, this will become a huge challenge.
Finally, in the most likely scenario, the pandemic peaks in various regions will differ in time and magnitude, and there could be second outbreaks. This would suggest that the effects of the pandemic on the economy will stretch out beyond the second quarter. Aside from the impact on demand, it also implies that a supplier’s factory may be running one time period (for a few weeks) and then closed the next period. Because of the variability in timing, companies should be able to better utilize resources by moving capacity allocations from highly affected regions to minimally affected ones. In such a scenario, we are likely to:
- Reconfigure supply chains, since some critical but vulnerable suppliers will go out of business.
- Reposition inventory in anticipation of certain regions or facilities being under quarantine.
Given that a shock of exceptional magnitude is already happening to the worldwide supply chains, both on the supply and on the demand side, it is crucial for companies to prepare for each recovery scenario.
For this purpose, apply some of the concepts and steps discussed in “Managing Unpredictable Supply-Chain Disruptions,” a 2014 Harvard Business Review article that gives companies ideas for riding out a storm such as the current one. Specifically, it is focused on the following objectives:
- Identify exposure to risk associated with parts and suppliers.
- Prioritize and allocate resources effectively.
- Invest in mitigation strategies such as booking logistics capacity.
To achieve these objectives, the paper introduces and applies the following concepts. A central feature is time to recover (TTR), the time it would take for a particular node — a supplier facility, a distribution center, or a transportation hub — to be restored to full functionality after a disruption. For instance, in a recent survey of 3,589 Chinese suppliers in the automotive industry, 53% reported that they will be back to normal capacity at the beginning of April. Similarly, if the quarantine time frame dictates that a supplier will be down until the end of April, then this provides an estimate of the TTR. Evidently, TTR is scenario dependent, but this exercise allows the supply chain to identify what is likely to happen in each scenario.
The second concept is risk exposure, an assessment of exposure to risk based on a model that applies suppliers’ TTR to estimate the performance impact, including operational (lost production) and financial (lost revenue and profit). By combining suppliers’ TTR information with the details of the supply chain, including a product’s bill of materials, volume, and profit margins by product line and pipeline inventory, the method identifies the risk exposure associated with a disruption in each site of the network. This is done by simulating (and optimizing) the company’s response to a disruption at a specific site for the duration of TTR.
In this process, the model allocates available resources during TTR to minimize losses and identifies how much logistics capacity the supply chain needs in order to move inventory and components so that they are available at the right time and location. One outcome from such a model is knowing how long the company needs to shut down its own plants and when to restart building up capacity again.
This model has been used by Ford since 2013. It was initially developed to help companies identify risk mitigation strategies — in particular, to pinpoint the weak links in the supply chain. Interestingly, Ford has already applied the model in recovery operations to identify an effective way to allocate resources after a disruption.
The current pandemic, however, raises another challenge, that of blocking supply chain risk mitigation — that is, downstream supply chain facilities may stop upstream suppliers from delivering product to them. Indeed, although China manufacturing is recovering, it now faces a massive problem of dramatic decline in demand due to the shutdown in major markets. Specifically, Chinese manufacturers are faced with a second shock wave associated with canceled or delayed orders from all over the world.
Of course, there are numerous case studies describing how companies have recovered from a disaster like the 2011 tsunami in Japan or the 2011 flood in Thailand. But now we are facing a more severe challenge because of the pandemic’s prolonged impact on the economy in general and supply chains in particular.
Following this approach, I suggest five steps that supply chain executives should take to develop an effective recovery plan for their business.
- Step 1: Identify suppliers in affected regions and estimate TTR by scenario.
- Step 2: For each scenario, estimate demand and assess which products and assembly facilities will be affected by these suppliers and for how long.
- Step 3: Use the insight from the previous step to determine when and for how long you should shut down, or significantly reduce, manufacturing activities.
- Step 4: Determine how to ramp up capacity by focusing on sales and operational planning. Allocate the available capacity and inventory only to products that allow you to achieve your specific objectives during the recovery period.
- Step 5: Book logistics capacity as soon as possible.
The first step emphasizes that it is not enough to focus on strategic suppliers! As the 2014 article demonstrates, small suppliers that provide low-cost components may be even more crucial, as a short supply of certain components might force production shutdowns. Therefore, this step is really about supply chain mapping, which requires connecting with tier-one suppliers and identifying their suppliers and, if possible, their suppliers’ suppliers. This step needs to be done immediately. A few technology companies can help accelerate this process (see, for example, Interos or Resilinc).
This step also requires the team to agree on the various scenarios to analyze. While I recommend the three possible scenarios mentioned earlier, you’ll want to consider others, of course.
The second step requires the company to estimate demand and assess risk exposure. One way to estimate demand is to apply statistical forecasting (or machine learning) techniques and use external data, in particular, recovery information from South Korea, Hubei, China, and Henan, China. To assess risk exposure, evaluate each scenario by applying the risk exposure model described earlier to understand the impact of disruptions in various regions on products and assembly facilities. Importantly, the model helps managers identify which suppliers/regions in the network create the greatest risk exposure — often highlighting previously hidden or overlooked areas of high risk.
The third and fourth steps apply the same (risk exposure) model to determine the best response to a node, or several nodes, being disrupted during the TTR duration. In particular, they allow the company to compare the costs and benefits of various alternatives for mitigating impact and hence identify the best allocation of available resources to finished goods. Specifically, the fourth step suggests which finished goods to eliminate from your offerings for the time being and what portfolio of products to focus on right now. For example, in the automotive industry, one would expect original equipment manufacturers (OEMs) to focus on various types of emergency vehicles because government demand for these vehicles will likely spike while auto sales to consumers are likely to decline.
If the third, most likely scenario is realized, companies should benefit from peaks in different regions happening in different times, since this enables better utilization of resources. To take advantage of peaks and valleys, the company will need to run the model on a weekly or monthly basis to identify how to move capacity around.
In this process, you may want to consider creative solutions to critical components and suppliers identified in the analysis. They could include paying more for certain components from alternative suppliers, financially helping out small critical suppliers, or using different materials, if applicable. Alternatively, you may want to share blueprints of certain critical components with suppliers that can modify existing tools and machines, enabling production of these components.
The impact of the analysis described above is a clear plan that details:
- When and how long supply chain facilities will be down.
- How fast to build capacity.
- Where and how much logistics capacity to reserve.
- What product to focus on.
The implications for consumers of this analysis should be obvious. Not only will consumers likely face shortages and longer delivery time for many products, but they will also face smaller product variety. For industry, the expected decline in sales and revenue in 2020 will likely force companies to implement significant cost-cutting measures everywhere.
In the long term, anticipate a significant supply chain restructuring. The U.S.-China trade war already motivated many industries to start the process; the pandemic is likely to accelerate it. Indeed, fashion retailers will likely move manufacturing to countries like Vietnam, Cambodia, and Malaysia. Meanwhile, high-tech companies will move closer to market demand, turning to Mexico and Brazil to serve demand in North America and Eastern Europe for the EU.
More importantly, companies need to improve the resiliency of their supply chains to handle global, not only local, disruptions by applying some of the concepts discussed and by achieving visibility into their supply chain and, in particular, their pipeline inventory.
The author would like to thank a few colleagues who helped shape this article. This includes Pierre Haren, Will Ma, Narendra Mulani, and Chung Piaw Teo.