First came the COVID-19 related shortages of masks, ventilators, and tests.

Now, experts are warning about a critical global shortage of glass vials needed to bottle billions of doses of a coronavirus vaccine that, some say, could arrive by year’s end. Not only are the glass vials made of a rare sand, they are difficult and time-consuming to produce.

It’s the kind of supply chain management quandary that Naren Agrawal, the Benjamin and Mae Swig Professor of Operations Management and Information Systems, has spent the last 30 years researching.

Think of supply chain management as the orchestration of all the activities involved in the creation and flow of goods and services from the raw material stage to the finished goods stage. The goods are transported to sellers, and ultimately purchased by consumers. 

Done well, the intricate process works efficiently, minimizing costs across all partners in the supply chain. Throw in a curve ball like a pandemic, or a natural disaster, or a war, and it’s hard to find flour or buy certain meats at the supermarket. 

We asked Professor Agrawal, whose career has focused primarily on companies in the retail and high technology industries, about lessons the supply chain management profession can learn from the COVID-19 related disruptions.

It’s been several years since the Fukushima earthquake in Japan, and two other deadly coronaviruses that surfaced in countries around the world: the Middle East Respiratory Syndrome (2012) and Severe Acute Respiratory Syndrome (2003). Each time, weaknesses in the global supply chain surfaced. Are we just re-learning the same lessons?

What we are going through does bring back painful memories. The loss of life and property was obviously enormous and highly regrettable in each case. But there are some big differences this time around.

For the U.S., the key impact of these events was supply side disruption. For example, when the Fukushima earthquake occurred in 2011, Japan was responsible for the production of 20 percent of the world’s computer chips, 60 percent of silicon wafers, and 90 percent of BT resin for circuit boards. Supply shortages of these parts severely impacted the production of cell phones, computers, and electronic components used in automobiles. 

But this time around, the disruption is a supply side and a demand side disruption. With the entire world locked down, rising unemployment, and serious concerns about job security, the demand has been seriously affected as well.  So, we have been hit by a double whammy. Challenges in the distribution sector, and concerns about employee and customer health have further exacerbated the outcomes.

Secondly, the impact is truly global this time. You could argue that the impact was relatively contained in the other examples. In today’s highly connected world, our recovery depends also on the recovery in other countries.

In the same vein, experts predict after COVID-19, some companies will invest in back-up supply networks, but others will simply hope for the best, and do little or nothing. Really?

Companies, as well as individuals, tend to be heavily influenced by “recency bias.” With time, some of these lessons will fade away. At a personal level, for example, dentists tell us folks tend to be diligent right after a checkup, but they start slacking soon thereafter. The same is true for companies after major disruptions. So if they don’t respond right away, they may not in the future. 

A second reason is that decision makers tend to underestimate potential losses from such low probability events. This is why households tend to under-insure even in flood prone areas, and companies may choose to be under-prepared to handle crises. 

But, ultimately, it boils down to an exercise in understanding and quantifying the tradeoff between the perceived likelihood of a disruption and the impact of such an event.  

Hedging against various types of risks does require an investment of some sort—inventory, excess capacity, information systems to provide timely information, backup sources of supply, etc. These investments might be perceived to be counter to a company’s efforts to be lean and efficient.  

Unless the C-suite makes risk management a key element of a firm’s overall strategy, they are likely to find themselves in a similar situation in the future.  

What advice would you offer businesses in the aftermath of the pandemic?  

Designing an effective risk management strategy is critical for supply chain resilience.  This requires companies to identify key risk categories relevant to their supply chain, pinpoint the key drivers of each kind of risk, and develop a comprehensive strategy to mitigate risk, recognize the event when it first occurs, respond to the event, recover from it, and learn from it so that they are able to handle it better the next time around. 

Unfortunately, this is challenging because the risk categories span a very vast range. These range from routine uncertainties in supply and demand, to occasional disruptions such as cyber security events (hacking of Target’s credit card number breach), quality mishaps (infant formula contaminated by melamine), natural disasters resulting from earthquakes (Fukushima), fires (Rana Plaza in Bangladesh), and diseases (SARS/MERS/coronavirus), to disruptions due to political events (wars). 

An information system that can alert companies to the relevant information as events occur is critical. As supply chains evolved globally, they became much more complex. While companies may have a very good idea of who their immediate, first tier suppliers are, they may have little idea of who their second, third or fourth tier suppliers are. So the first step is simply developing an appropriate map of a company’s extended supply chain. But this is not enough. They also need to understand the vulnerability of the deeper tier suppliers to various risks, and, in real time, have visibility into risky events and disruptions, which would allow them to respond quickly. 

For example, if there are floods in Thailand, it would be critical to know exactly which suppliers, or their suppliers would be affected, how much supply would be affected, and when would that impact be felt. 

While a very large percentage of North American companies claim to engage in different aspects of digitalization of their supply chains, the fact is that at this stage, the majority of these companies have adopted a piecemeal approach to digitizing discrete steps in their overall processes. I think that we are some ways away from comprehensive digitization of supply chains.

In addition to such information systems, companies also need to design resiliency in their supply chains, and invest in flexibility. For example, strategic stockpiles of inventory may be required in some instances to build resiliency and flexibility, though there are other strategies as well. 

China had to shut down or slow down many factories that produce consumer goods, not to mention processors and other chips for companies like Apple and Microsoft. Are consolidated centers of production a good idea?

There is certainly a lot of talk about consolidation, and onshoring manufacturing operations, but it’s not that simple. Consolidation provides the benefit of economies of scale, while the key driver of advantage from onshoring or nearshoring is reduced lead times, and perhaps a move away from what is perceived to be a risk prone area currently. However, while there might be some cost or efficiency advantages, either approach doesn’t necessarily buy us supply chain resiliency.  

What if we consolidate all operations in one factory in New Orleans, and we have an unfortunate repeat of Hurricane Katrina? Therefore, we need to consider a more comprehensive risk management strategy toward designing and managing supply chains, that includes factors such as economic conditions, production cost differentials, availability of appropriately trained workforce, taxes and tariffs, access to leading edge technology and innovation. 

Regulatory factors are also very important; foreign countries may require certain activities be performed locally as a pre-condition to allowing access to their markets for our products.

Often, the optimal strategy is to adopt a portfolio approach, much like we would for personal financial investments. In other words, the goal is to create an appropriately diversified supply chain that is tailored to specific instances.

“Optimal” supply chain management means increasing efficiency and removing bottlenecks. Are employees considered bottlenecks?  

Absolutely! Human capital is one of the key inputs needed in any production system, whether for goods or for services. While in some industries, this resource may be very easy to add or subtract, it may not be the case in others. Currently, front line workers like trained doctors and nurses are critical to our ability to successfully deal with the pandemic, but they’re a major bottleneck in the U.S. Since it is not possible to suddenly increase our supply of trained doctors, hospitals and medical facilities have had to be extra careful in deploying doctors and nurses, and have struggled to avoid infection, which would deplete an already scarce resource. 

We have also seen innovative approaches to managing bottlenecks in human capital. For example, in Germany, out-of-work waiters were recruited to harvest fields, and in the U.S., companies like California Giant Berry Farms have utilized workers displaced from the construction industry on their farms. Lineage Logistics, a U.S. based logistics company, shifted salaried employees to warehouse roles and is recruiting laid off restaurant and service workers from temp agencies. At homes, grandparents and older siblings are taking care of children during work hours 

Robots can’t get coronavirus. Will the pandemic make them even more attractive to companies?

Great question. The contagiousness of the coronavirus has provided a boost to novel applications of robotics, and this momentum may continue.  

In warehouses, like at Amazon and Walmart, robots were already used to improve efficiency. The COVID-19 outbreak has caused both companies to explore further use of robots for sorting, shipping, and packing. 

In China, robots were deployed at a Shenzhen hospital specialized in treating COVID-19. These robots were typically used in retail and hospitality scenarios, and were modified to perform tasks in hospitals—providing video conferencing services between patients and doctors, monitoring body temperatures, and disinfecting designated areas. Walmart is using robots to scrub floors.

Fast-food chains like McDonald’s are testing robots as cooks and servers. My favorite example is Spot, a dog-shaped robot developed by Boston Scientific, enforcing social distancing in public areas in Singapore.

Obviously, a key factor preventing wider use of robots is cost. In order for these products to become cost effective, we need scale. Wider deployment of robots will also require further advances in AI, computing and communications technologies. 

Can you talk about your current work related to COVID-19?

I am currently involved in three projects that focus on challenges heightened by the pandemic. 

The first involves advising a molecular diagnostics company that makes equipment and tests to identify and analyze genetic disorders and pathogens, including tests for COVID-19, on strategies to improve the efficiency of their manufacturing process. 

The second is a collaborative effort to develop a new, machine learning-based paradigm for supply chain planning that leverages big data, cloud computing, and advanced optimization techniques to help companies respond to disruptions in an agile manner. Our method allows supply chains to respond quickly to changes in their environment, rather than rely on slow and inefficient planning processes based on flawed forecasts. Our initial work with companies in the semiconductor equipment and consumer electronics industries has been very promising.

The third initiative is with Give2Asia, a U.S.-based public charity founded in 2001 that facilitates philanthropy across international borders. Since the outbreak of the pandemic in January 2020, we have facilitated more than $17 million in donations and commitments to fund local COVID-19 disaster response projects covering 23 countries. As one of the few U.S.-based nonprofits registered to support NGOs in China, the organization facilitated emergency grants to more than a dozen charities working in Hubei Province.