Publication Date
September 17, 2026
When disaster strikes, the response is far more complex than mobilizing aid and supplies. In this episode, we explore how supply chains function as critical infrastructure in humanitarian crises and why understanding them as interconnected systems of systems is essential to saving lives and rebuilding communities.
Joining the discussion is Dr. Jarrod Goentzel, founder and director of the MIT Humanitarian Supply Chain Lab at the MIT Center for Transportation and Logistics. Drawing on over two decades of real-world crisis response experience, from the 2004 Indian Ocean tsunami to the Liberia Ebola outbreak to post-wildfire housing reconstruction, Dr. Goentzel reveals why single-point interventions fail and how systems thinking transforms emergency response. The conversation spans four critical domains: the port bottlenecks that reveal hidden supply chain dependencies, post-disaster housing as an industrial manufacturing challenge, medical supply chain surge capacity during pandemics, and fuel as the foundational commodity enabling all disaster response. Dr. Goentzel also discusses the practical opportunities within the 21st Century ROAD to Housing Act for building resilience by design and how communities can collectively demand supply chains that meet human needs before the next crisis.
Transcript
- Welcome to another episode of "Supply Chain Frontiers," the MIT CTL podcast where we explore the trends, technologies, and innovations shaping the future of supply chain management. I'm your host, Mackenzie Berry. Today, we're diving into the work of the MIT Humanitarian Supply Chain Lab at the MIT Center for Transportation & Logistics with the lab's founder and director, Dr. Jarrod Goentzel. So glad to have you on the podcast today, Jarrod.
- Yeah, thanks for inviting me.
- Can you introduce yourself and why you started the MIT Humanitarian Supply Chain Lab?
- So actually, I came to MIT to help with some educational innovation programs. We set up our first global SCALE center in Spain. That was my first role, to help create a new educational research platform in a different country, similar to what CTL does. And we've had many more of such centers now. When I came back after doing that, I worked on the master's program. I ran the master's program at MIT for a couple years. But all that time, I was looking at what research I wanted to dive into. And I had been doing research with companies like CTL mostly. A lot of our work is driven by industrial applications. But I grew up responding to disasters in Kansas. So I responded to tornadoes. My church and other organizations would be the ones to show up after a tornado had destroyed a house and help them muck out the basement and so forth. And I remember loving the energy of showing up after someone's had their worst day. So that was in me. The first year I was at MIT, there was a student who had worked in the Peace Corps, Tim Russell, and he said, "Hey, I kind of would like to look at humanitarian supply chains." It was not a field at the time, I'll be honest. It was not a study in academia, but I was interested, and we started doing research. Then what happened in December that year, this was 2003, there was the Indian Ocean tsunami, and that was a tsunami of the world realizing, "Hey, logistics and supply chain is pretty important after a disaster." And so there's a whole study that emerged following that big crisis. In fact, the UN organizations, Red Cross, all changed how they approach things based on that crisis. But one of the big things is that logistics and supply chain became more of a profession in this space. And so we kind of rode that wave. And eventually, in 2010, we helped be part of a larger response to the Haiti earthquake on campus, which really then hooked me on, "I need to focus more time on this research." And so in 2011, I started the lab. And actually, Tim Russell, who helped me start thinking about this as a student, has come back to work with us. He's been working with our lab now for over 10 years. So it's kind of a full-circle moment that a student came with an inspiration. I had an interest, and we together started exploring this topic.
- I love the story that you told too of your initial interest in disaster response because it seems like a no-brainer now, where you have a disaster and you need a communal response. You have to have a communal response, and supply chains are the mechanism for that. So your lab works across four areas: supply chain resilience, housing innovation, health emergencies, and humanitarian action. How do these areas connect for you all?
- So the mission statement we have is basically understanding and improving supply chains to meet human needs. So it all starts with a human need. What needs are there? And there are a lot, especially in a post-disaster setting. But in general, the idea of, "How do you have a supply chain react to a crisis?" is the connecting part. Initially, we started by doing aid supply chains. How do you set up a specific new supply chain to bring in aid because the markets are not recovering and not resilient? Especially internationally, that's a common situation because markets are more vulnerable internationally. But even in the US, we have a disaster. It takes a while for the private sector to get going again. And so the initial work was with aid organizations like the UN, the Red Cross, NGOs like International Rescue Committee, Medecins Sans Frontieres, and so forth. These are organizations that set up a specific supply chain to respond. Soon, actually, part of the Indian Ocean tsunami was a realization that if we give people cash, they can go to the market. If the market's ready and resilient, that's a better way to support human needs being met by a supply chain because they know that supply chain, it's in their community, and it has more capacity than setting up a bespoke supply chain. So we started doing research, and this is where Tim actually was part of this initial work with World Food Programme in Sudan on cash and vouchers. How do you stimulate a market, and how do you make sure the market's ready to respond? In our Center for Transportation & Logistics, we study private-sector supply chains. So the resilience of those supply chains, creating a market response, is where most of our research is. And that's where the resilience work comes from. Housing innovation is a specific need where we think there's a lot more opportunity for better resilience. So that's more of a domain, a particular kind of supply chain, construction supply chain to meet housing needs, is a big opportunity, and we're diving into that. Now, health emergencies, health is obviously a sector that has a lot of specialized knowledge, but the supply chains are still structured the same way. So we support health emergencies where there's been a lot of inquiry from the public health profession and medical doctors on how to make the supply chain work better. And then, just in general, humanitarian action, we're ready to support any need, and we just try to find the supply chains that are not working as well in meeting those needs and try to reach out and make those connections, and we always have to be invited in.
- Yeah, absolutely. So thinking about the larger systems at play here, your lab frames problems as systems-of-systems challenges. Why isn't it enough to just solve one part, like building faster or moving fuel quicker? Why do you have to look at the whole thing?
- Not everyone thinks in terms of systems and doesn't realize sometimes how many systems are at play in something that seems pretty straightforward. So you mentioned housing. Most people understand you need a builder to rebuild a house or repair it, but the builder needs cash short-term and long-term, so the banks have to get involved, and then the insurance company has to be involved. In terms of rebuilding, to make sure it's going to be meeting the code for a future disaster, you have architects and engineers involved. The builders rely on a building supply chain, which is its own supply chain. They have a construction process they have to do. And then there's also all the inspections and city planning. So there's a lot of things that are a lot of systems that are required to repair or rebuild a house. And sometimes we think, "Well, it's all just one thing, the builder's doing it," but the builder has to have all those other systems working together to enable them to perform the activity of repairing or rebuilding the house. And so that's the way supply chains work as well, as a system of systems.
- And I'm sure in your line of work, you have a lot of stories to tell over the years. Can you give a concrete example where failure in one part of the system stopped everything? What happened?
- Yeah, I'll tell one that people might remember from 2021, when the Port of LA seemed to be closed. All the ships were queuing up offshore, and they couldn't offload, the big bottleneck in Los Angeles. And you saw in the newspapers people trying to talk about how to unclog that bottleneck. And in fact, the government stepped in to say the port should be open 24 hours to unclog the bottleneck. And they did that, and nothing changed because that wasn't the bottleneck. So understanding the system is required to figure out what actions, what interventions are going to unclog or make the flow work, go faster. And interestingly, this is, and I give a definition of a supply chain, by the way, there's a textbook that said, "All of the parties involved directly or indirectly in meeting a customer need." And so freight transportation is a good example of lots of parties being involved. So you need to have the ocean carrier bringing the ship coming in. You have to have the port being able to offload the container, position it. There's all sorts of equipment in the port that have to position it to get ready for the drayage company. A drayage company grabs the container, takes it out of the port to another facility, like a transloading facility, where it might be put on a chassis to go to a warehouse. So that's already a lot of parties involved in that particular supply chain. So where was the bottleneck in 2021? It wasn't the port. It turned out it was the chassis that enabled the containers to be moved from the port out of the port. And this is something we had actually seen in Puerto Rico following Hurricane Maria, because the hurricane hit, and the port of San Juan was inundated with containers, but the warehouses didn't have electricity and so forth. So when the container arrived, they couldn't really offload it and so forth. So they let it sit as a storage facility, and therefore it tied up the container, the chassis underneath the container. And so the port got backed up because the containers couldn't get unloaded at the facilities, and then the chassis was tied up. So something similar happened in 2021, and one of the easy ways to unclog this was then changing a requirement such that empty containers in Long Beach, they can only be stacked two high for aesthetic reasons. The city of LA, of Long Beach with them, was able to change to stack four or five high so that empties could be brought out and the chassis underneath it could be freed up so it could pull containers out. You needed to know the drayage operators and the other people in the supply chain to be able to figure out that that was the real cause of the bottleneck.
- Yeah, and then, of course, all the communication problems
- Yeah.
- That happened during a disaster too that you have to work around.
- And it actually took, I mean, we had learned this, and thankfully one of the freight forwarders, a big company that does freight forwarding, they did a tweet storm to help the public understand this complex challenge that we see, the system of systems, as researchers, but sometimes the public doesn't. And so that communication of how the system worked, once they did that, he went around in the port and showed videos of all this stuff. Once people could see how a port worked, the system of systems, then you could say, "Oh, that's obvious. That's clogging that. If we free up that, that'll free up that, and then the thing can flow again."
- Pivoting to dive more into housing, which is one of your key focus areas. With climate change, the frequency and severity of natural disasters is increasing. We've had a lot of conversation around housing becoming more unaffordable, in part because of increasing insurance costs due to increasing disasters. Can you walk us through why post-disaster housing is a supply chain problem?
- It's interesting because I think a lot of people think about supply chains as delivering goods or items or commodities, but projects are also a supply chain. So construction is a project, but it's part of a supply chain. There's goods out to come in, but also you're building a house that's innately a manufacturing process. It's a supply chain activity. It's a production activity. And those aren't studied as much in academia for some reason. And because there aren't as many project supply chains in the marketplace, actually humanitarian response is kind of a project supply chain. And in fact, a lot of people are starting to make the connection to, "Why can't we build houses like we build automobiles?" Automobiles are kind of like a project supply chain, but, I mean, it's in a factory, so people don't think about it, but you have to have lots of things come together to create that big product. So it's more complex. So post-disaster housing, one of the challenges is that after the disaster, there's not a lot of resources to meet the surge in demand. There's an ongoing market for housing and obviously, as a lot of us know in the country, there's not been enough supply. When you have a shock, there's even more demand. And so that market has been disrupted. That's why we are looking at post-disaster housing as a supply chain problem, because the local conditions that are required for traditional onsite building are not there to meet the surge in demand.
- One survivor in your research spent 10 months rebuilding a home that took 45 days of actual construction. Can you take us through what's happening in those other eight and a half months, and what does it tell us?
- Yeah, that was interesting. We had a roundtable in May, and I hope we can share the report as part of this podcast. And the story's told in there, but it's interesting because it was actually, the participant was a home builder. So they know the sector. They know how construction works. And hence, the 45 days of construction went pretty well. The other eight and a half months were consumed by engaging insurance companies, the banks with the mortgage, doing the permitting. In post-disaster, there's a need for a lot of permits. And so the local authorities need a surge capacity to process permits and all the coordination steps it takes to get the ability to build. And so that builder had not anticipated that in a post-disaster context, things that work kind of normally in a non-disaster context can take a lot longer because each is a system, each has dependencies on other systems, and they all have to be orchestrated to make sure that you have the ability to start rebuilding and to follow through on the process. Surge demand can really disrupt an efficiently designed system, especially a system of systems. And so when there's a surge in demand, it's not designed to scale up, and that's what we're trying to look at.
- You talk about industrialized construction in the report as a solution, and you talk about also the terminology, how that can be confusing. When we say modular, manufactured, or factory-built, they can get mixed up depending on which audience you're talking to. Can you talk about why this terminology problem matters and some of the issues that it's caused?
- Yeah, you're right. This term, industrialized construction, is still pretty new, and it comes out of trying to address the fact that people most think about manufactured housing as being like mobile homes and trailer homes. And that's a particular way of constructing something that now, due to HUD in the '70s, can be a permanent dwelling, but modular is another way of building something offsite in a factory setting or something and bringing it to a site and living in it. But it's not the same industry as what we call manufactured housing. And so they get conflated, and it's really hard for consumers to know, well, what's a manufactured house? What's a modular house? So part of this is helping consumers know that you can have a house that's built in a manufacturing facility that is not a trailer home. In fact, ironically, some of the highest-end, most sustainable, energy-efficient homes in the country are built using modular approaches or manufacturing. So we've got manufacturing kind of manufactured housing is kind of on the two extremes, more affordable housing and the kind of mobile home space, and then high-end, carbon-neutral houses on the high-end space. And what we need to do is fill in the middle with manufacturing playing a role. There's real implications. So first of all, there's the confusion, right, you know, and so, of these different things for consumers, but also one of the consequences is setting policy with this kind of industrialized, modular, factory-built, system-built space for a while now. And different states have been trying to facilitate that to happen. So Virginia was very early in changing their codes so that modular construction could be a norm across all jurisdictions. It's actually a state law that the local jurisdiction can't set codes that don't allow for this. So different states have different policies, and the way they talk about it is important because the terminology, when you put it into statutory constructs, can make a difference. And I think a lot of times these terms are not well understood, and states are developing laws meaning one thing, but writing something else and conflating the terms. And then there's a lot of work later on to understand what was really meant when that law was written and then interpreted appropriately. And so there's a lot of corrections that have to be done to policies that are developed. And so there's real consequences in not having a proper scientific way to describe this industrialized supply chain of construction.
- Yeah, and part of what's in the report is standardizing, offering clear definitions for each of those.
- The first thing we had at the roundtable is agree at what terms meant because they're used differently in different contexts. And I'll even say there's a massive legislation this summer passed in terms of housing to make it more affordable, the "21st Century ROAD to Housing Act." Even that legislation still suffers from terminology that will have to be cleaned up as we go forward trying to implement it. And fortunately, then, this act actually has within it some calls for research for HUD to sponsor, working with their partner, the National Institute of Building Sciences, NIBS. They were set up in the '70s to help HUD make housing more of a scientific approach. And I think now is the time to think of not only building science, but also management science of supply chains to make housing more affordable by designing the industrial supply chain.
- Yeah, absolutely. And what can industrialized construction do that traditional onsite building can't, especially in a post-disaster context?
- So if we think about to build a house, and actually one reason I'm passionate about this sector is I grew up the son of a general contractor, so I was on job sites throughout my youth helping build houses. And so that typical model is, we ship a bunch of materials to the site, and then you have a general contractor who hires subcontractors who do different roles to assemble the house. And so everything is being brought to the site, but there's inefficiencies in that. And one of the ones that you'll hear most general contractors complain about is, "How can I get my subcontractor there on time to keep the project going?" because time is money, because they have taken out a loan to buy materials, and cash flows are really tight for builders. And so they want to get stuff done faster because they have a loan to buy all the materials to build the house that they're paying interest on. And so time is really important, but it's difficult to orchestrate the production of a house when you have to bring the workers to the site. And you have to bring the materials to the site and assemble it in conditions that aren't ideal. Instead, we could bring the materials to an offsite location, and you don't have to build the whole house, you build a room or even just a part of room, like a bathroom. In a disaster setting, you can do it in a place where there has not been a disaster, where there's not suddenly a demand for plumbers and electricians and carpenters. So the labor market is more stable far away. Second, they can build in an environment that is more conducive to construction. And third, this is one of the big things, is we can think about when we start centralizing and doing things consistently in one place, we can bring in industrialization. And that really means, like industrial engineers, that's my field that I got my PhD in. Once you have a really defined process and a facility, you can really tweak it to make it better. And there's economies of scale and scope that can be realized. And that can make it more affordable. And also, quality control comes into play. You can now do quality control in the facility and not have to rely on an inspector going out on a site and saying, "Hey, did these people who came out one at a time build this the right way?"
- Right, and I imagine there are different rules, of course, for each state, so then there's a lack of standardization there as well.
- And you bring up something there that's really important. And one question is, "If it's so obvious to build offsite and there's a lot of efficiencies, why hasn't it happened?" One reason is because there's not consistent demand. Once you have an industrialized process, you need a more standard product. And if every product is built differently, that's the way our system is built. Each jurisdiction, even local jurisdictions, can say, "This is how you have to build the house. You have to build it this way." And then we send an inspector out to go look at it and make sure you built it right. And that's the quality assurance. Whereas we could flip this around and say, "The code is performance-based. It should just do this." We then inspect the factory to say, "Do the processes in the factory provide the quality necessary to perform as we want it to in our jurisdiction?" And then the quality assurance is done through inspections randomly at different times just to make sure that the process is still delivering the quality that's needed. Now we're not inspecting every house to see if it's built right. We're inspecting houses as they come through the production facility to say, "Is this module built right?" And there's still going to be some onsite work to put things together. And that's one of the questions, is what do we do onsite versus offsite? That's a big research question we're diving into. But when you can have most of the quality assurance done offsite to a performance-based code that could be standardized more, then the whole market can emerge. And right now, it's up to each jurisdiction. Each city or town or county can say, "Build it our way," and the factories can't serve that many different ways of building.
- You emphasize in the report that resilience must be made generally accessible rather than a one-off. What does it take to make resilience the default design?
- Yeah. Kind of building on what we just talked about with the inspections. So, for example, in our roundtable, we focused on how do we not only build faster or cost-effective, really, how do we build better, especially in areas where fire hazard is real. And you could say to people, "Well, don't live next to forests that burn," but that's their home. It's their livelihood. But we can also build homes better, right? And especially if we had a whole community building homes to a higher standard, then there's a risk pooling that happens. If the homes in Altadena now were rebuilt to a resilient standard, then the fire that might come down the mountain in the future is like a firewall for Pasadena. We can build better to then allow ourselves to live in places that have increasing risks, but still have resilience because we've built them better. But to do that, we have to have a standard way of building that is going to provide that resilience. Right now, the ways we build for resilience are through specific program certifications. There's one focuses on wind damage and hail damage for building roofs. It's called a "FORTIFIED Roof." That's a certification that the insurance companies have seen. Now, that makes a difference. It makes the house more resilient, but it requires, again, an inspector to go out and make sure it's built right, and not every homeowner chooses that. If instead we said "FORTIFIED" is now part of a production process in a factory, it can be done, first of all, less expensively because it becomes the default way of building in that factory. And second, the quality assurance is guaranteed so that every house that's rebuilt has it. So if only we had had a factory ready when the LA wildfires happened that could build, and there's an emerging standard called "Wildfire Prepared," if we had a "Wildfire Prepared" factory ready to go and all of the other systems of systems ready to engage that factory, we could rebuild Altadena in a way that is resilient for the future. And that's the vision we're trying to play out.
- This episode is brought to you by the MIT Humanitarian Supply Chain Lab at the MIT Center for Transportation & Logistics. The lab designs supply chains, develops planning approaches, and facilitates cooperation across public, private, and nonprofit sectors to meet the needs of humans in crisis. Learn more at humanitarian.mit.edu. So pivoting to the health emergencies and preparation part of your work, your work directly supports health systems to surge medical supply chains during emergencies. How is surging a medical supply chain different from, say, surging fuel or surging housing?
- Yeah, well, first of all, medical supply chains, there's a different kind of quality standard that's needed for human health, and appropriately so, right? And so that has implications on the supply chain. For example, for treatments that require cold chain or refrigeration, your supply chain has to be designed completely differently to make sure that happens. And also, there's just the quality of the interventions that the supply chain is supporting has to always be there, but also the timeliness, right? In thinking about medical treatment, you need the item to be there when the patient is there and the practitioner's there to administer the treatment. And so medical supply chains are just a special version of making sure the supply chain is having quality and timeliness appropriate to a medical intervention.
- What are some of the examples of the health emergencies that you've been involved in where you've had to apply this work?
- When we're invited in, and any of our humanitarian actions where we're responding and being active part of a response to a disaster, we are invited in. But especially in the medical space, we need medical professionals who need our supply chain and systems engineering. And so one example is in 2014. I have colleagues in Boston who were working with the Liberian Health System to set up after the war, their first reset up there, establish their residency program to train more medical doctors, and then Ebola hit. And so they asked for our support in helping provide some supplies. Then the Ministry of Health said, "Hey, in addition to providing for treatment centers for Ebola, our hospitals need protective supplies so that they can do regular procedures, especially like childbirth, which has a lot of blood exposure, and that's how the pathogen is transferred through blood." And so the doctors were invited by the Ministry of Health to set up a supply chain to support the public hospitals for regular protective equipment for regular procedures, and then they asked us to help set up a supply chain. And that was a particularly challenging time to set up a medical supply chain with that kind of protective equipment because there was such a spike in demand. And there's a whole set of stories on how we had to be creative to respond to that emergency. But that's an example of, with the right medical invitation, our supply chain knowledge can be really useful in setting up an intervention to make sure that the nurses and doctors in Liberia providing regular procedures were also safe while there was an Ebola epidemic.
- Can you talk about some of the preparation work that goes into working with health systems and medical supply chains?
- Yeah, actually, and one of the things, kind of like what I was saying in Liberia, they didn't have the protective equipment. And so one of the ideas is, well, "Why don't we store more of that?" It's a common question. In fact, that came up during the COVID pandemic, is, "Why don't states have more stockpiles to support the hospitals for things like N95s?" I think everyone has some connection with that supply chain. The difficulty of stockpiles are a challenge to maintain because, first of all, you have to have a capital investment in things that you don't necessarily need. And second, they might expire. And so you first have to buy a lot, and then it might go bad, and so you don't want to take the risk. So how do we think about stockpiling for a pandemic or an epidemic? We actually got involved through Mass General Hospital during the pandemic with the state of Massachusetts and the regional states, actually, on how would we set up a stockpile more effectively. And there was a whole study on that that we did to talk about how you could assess what kind of stockpile you might need for a future epidemic that would be similar to, or a variant of, the COVID. Then a few years later, they came back and said, "We love that report. We now have set up a stockpile. Can you help us figure out how to maintain and make sure we have the right items?" And so we just finished a project with the state of Massachusetts on establishing rules for deciding what to buy and put in their warehouse, in their stockpile. And it's interesting because it applies a supply chain concept that is common in academia. We teach it in all of our classes, called the newsvendor problem, but it's difficult sometimes to formulate that problem in a way that you could use it practically to say, "Well, how do I make sure I don't buy too much or too little, given the risks of, if I buy too much, it might expire, or too little, I might not have enough in the crisis?" And so we helped them develop a more practical tool to create the data and kind of create synthetic data to make that newsvendor problem something we could apply directly based on the incentives and the logic that they understood that the health authorities wanted to be able to provide in the state.
- So you would say that now we're more prepared, should another pandemic occur, God forbid.
- That's the idea. And there's actually two things, to be honest, that we've given them because in the first project, we helped them figure out how to quickly assess what demand would be based on a trajectory of a particular epidemic. Second one, this recent one, is, given that, how would you then stock ahead to show that if they had had a stockpile prior to the COVID pandemic, how many weeks they would've had to be able to adjust and find a source of supply for things like N95s. I mean, stockpile's basically buy you time, but that time could be essential because you're going to have to adapt your supply chain.
- Yeah. You all work across many different kinds of disasters, which I imagine makes your work quite challenging. You have to be very adaptable and flexible. What have you learned about preparation that can actually transfer across different types of emergencies, whether it's a hurricane, an earthquake, a pandemic, or other disruption?
- Well, talking about stockpiles buying time, right? It doesn't matter what the hazard is, hurricane, earthquake, pandemic. I mean, they'll have different characteristics. Sometimes you get a lead time. With a hurricane, you have some anticipation. So you could start doing some of the preparation work in kind of real time as you see it coming, right? And that's actually... So in those kind of situations, if you're preparing for hurricanes, you may not need as much of a stockpile. Because you can see it coming, and that adaptation and that getting ready for the hazard can be done in part by just mobilizing really quickly and moving items into position. But for an earthquake, you can't. And so then for earthquakes, you do need a stockpile. It's a lot about understanding the system of systems, which we talked about before. And how can we, regardless of what the hazard is, pandemic, earthquake, hurricane, the issue's going to be that somehow the supply chain and the systems are not able to meet demand effectively because of demand surges and because supply chain capabilities are disrupted. Warehouses might be out, roads might be severed, and so forth. So we more look at the types of ways that the disruption hazard manifests in the system, and then we prepare that way. So we're really preparing across all kinds of hazard types by just understanding demand and supply systems in a better way.
- Looking more largely at your humanitarian and global work, as you said, you've worked on several continents with development and community organizations in direct action. How does working globally change what you learn about resilience?
- Yeah, we have had the privilege of working on several continents, and sometimes it's interesting to realize the country I know best is the US. I know the government and the industry settings best, but sometimes working out of that context helps you see things differently. And we're trying to do something that we learned from the international sector, from the years of working internationally in disaster response, from something called the Logistics Cluster. So this is part of a humanitarian orchestration system that emerged out of that tsunami in 2004. That tsunami response is called the cluster system. So different medical specialty, different specialties like medical, water and sanitation, food, they have sectoral focus on how to meet people's needs, but there's also a Logistics Cluster on how to move all the items to support the food needs and the water needs and so forth. And so that Logistics Cluster is something, is an organization we've worked with for years, done a lot of research with them. And some of their approaches to how to orchestrate the supply chains that are trying to respond after a crisis are things that have not been fully adopted in the US by the national disaster management, the FEMA, or state disaster management organizations. And partly because they haven't been exposed to them, but in part because a lot of the supply chains that are needed to respond are private sector. And so there's just sometimes a difficulty in being creative between public and private sector to solve a problem. And that's where academia can play a role. And so we're trying to bring some of those concepts on how internationally they mobilize logistics resources for the common good in a community and apply them to communities in the US with the local government and the local retailers and health system providers who are all working together within a community to do something similar.
- So looking across disasters, you mentioned in a December 2024 roundtable on fuel supply chains that fuel is the number one concern in disasters because it enables everything else to happen. Can you speak more to that and why it matters so much?
- In a lot of humanitarian and emergency management organizations, they don't really explicitly plan for fuel. And it's always surprised me because in a post-disaster setting, fuel enables you to move things, trucks and ships and everything, but also you often lose power. And so electricity relies on generators that rely on fuel. So it is the essential commodity for everything else to work. I think it's because it's not something you often give directly to people who have suffered. Obviously, they'll be seeking fuel for their own needs, but they kind of miss out on the fact that if we don't have fuel, then we don't have the ability to respond. We got really involved with FEMA in enabling the private sector during their 2017 hurricane season. And when given the opportunity, they sponsored a National Academies study to see, "How do we think about supply chain resilience?" We decided at MIT to focus on fuel because that enables all the other supply chains. And we have learned a lot about how fuel supply chains work in the US over the years and kept doing work with FEMA on understanding how the upstream supply chain of extracting oil, refining it, getting it to a terminal, is often very separate and decoupled from the delivery to a fuel station to make it available to the public and the emergency responders. And so we've tried to close that loop in understanding that delivery, that last-mile part of the fuel supply chain, in a much better way to make it more resilient to surges in demand because you're going to get a surge because, again, fuel is the basis for everything in a response.
- I remember in the workshop, one of the participants mentioning that one of the big variables in disaster responses that you can't control for is human behavior. I remember him saying when people would see the fuel trucks coming in somewhere, following it with guns and waiting to be first in line, for example. How do you prepare for things like human behavior in a disaster context?
- Well, you can't fully, right? Part of it is experience and having seen it before, thinking through, for example, the thing you mentioned is that a lot of times they have the provision for escorts to bring fuel trucks to stations for two reasons. One is get some access to the station because if there's a line of people trying to get fuel, it's hard to get the truck in. So they make sure that there's a way to get the truck into the station. And secondly, if the truck arrives, that the surge for demand does not cause behavior to disrupt the fair allocation of that fuel to societies. But the bigger thing is, one thing you can't control, for example, in fuel, we did a big study with FEMA following, in 2021, there was a ransomware attack on the Colonial Pipeline, which distributes most of the fuel on the Eastern Coast, long story short. And so news got out, and so everybody, human behavior, panicked and went to the station to go buy fuel. So it turns out during that whole Colonial Pipeline ransomware attack, the terminals that had fuel near to the stations never ran out. But there were shortages and long lines at stations because people, again, behavior, went to go buy more than they needed. So if the word hadn't gotten out, no one would've noticed that disruption. In fact, there had been disruptions to the Colonial Pipeline of that duration in the past that did get publicity, and there was no issue in the station. So part of it is, you have to anticipate that in a disaster, human behavior will kick in to make demand really surge. And that's where we have to think about supply chains that might be able to respond to that human-induced demand surge that makes the supply system fail.
- We saw a lot of that in the pandemic as well, with the empty shelves. Looking ahead for supply chain professionals listening, what's one thing you wish they understood about building resilient systems that might change how they think about their work?
- We did a roundtable actually in the spring of this year, in 2026, on supply chain resilience and why it's talked about a lot now, but how do you actually get a company to invest in it? People say, "Well, we're investing in something we may not need." And how do we justify that to our shareholders if you're a company, for example, or to the public? If you're a government agency, you spent money on something that you didn't need, so. And what I often say in response is, if you're trying to understand resilience, but how you would make your system react and respond to disruption, you're learning about your supply chain. You have to know more about your supply chain to figure out how to make it adapt when disruption happens. When you learn about your supply chain in that way, you'll find a lot of spillover effects because you now know how your supply chain works better because you've tried to stress-test it. And by stress-testing, you can find opportunities for innovation. And I think that a lot of companies don't realize that investing in resilience and understanding your supply chain better so it can adapt better when needed gives you opportunities to reinvent how your supply chain works day-to-day. And that spillover effect is, I think, vastly underestimated by a lot of companies. We would love to work with companies who want to take a chance on investing in resilience and seeing what kind of spillover effects happen, because I guarantee that there's going to be a lot of day-to-day benefits come from understanding your supply chain better.
- Yeah. It seems like all of us can count on disruptions of one form or another.
- In fact, at that roundtable, one thing I talked about is another way to position investment in resilience is like buying insurance. We buy insurance we hope we never have to use. And so we also played with that framing internally in the company, is, if you know your supply chain really well and you can adapt better and faster to meet customer needs, that's the kind of insurance that your CEO wants to make sure that the quarter-to-quarter financial results are going to be good. And so instead of seeing it as an investment we might never need, it's insurance to make sure that our supply chain always performs like it needs to.
- And just one last question to close: what gives you hope about solving these problems?
- We're leaning much more into more direct engagement with communities to help shape the supply chains that serve them. So in our work with housing, I think a lot of this is, how do we help local leaders and local developers and local builders and local insurance agents and local banks all say, "Hey, if we all together said we want to build houses differently," then the supply chain might be set up differently to meet our housing needs, right? And I think if that kind of community action to mobilize demand for an innovative supply system can happen, then I think we will get better results in society. And so I think that can happen with housing. We're also hoping to do this with community leaders, like working with food banks and grocery stores to say, "Hey, how do we feed people after disaster?" If we talk together, maybe we can collectively have a better supply chain to respond to meet people's needs if there's a need for that. So I do believe that community-led design of supply chains is something that can have a material impact on disaster response. And that preparedness, again, it's hard to sometimes think about preparing for something that you hope doesn't happen. But I do believe even if communities say, "How do we feed ourselves in a crisis?" They'll think better about how to feed themselves day-to-day and meet human nutrition needs on a regular basis.
- That wraps up this episode of "Supply Chain Frontiers." A big thank you to Jarrod Goentzel for sharing his expertise and insights. "Supply Chain Frontiers" is recorded on the MIT campus in Cambridge, Massachusetts. Our sound editors are Dave Lishansky and Danielle Simpson at David Benjamin Sound. And our audio engineer today is Kurt Schneider at MIT Audiovisual Services. Be sure to check out previous episodes of "Supply Chain Frontiers" at ctl.mit.edu/podcast or search for us on your preferred podcast platform. I'm Mackenzie Berry. Thanks for listening, and we'll catch you next time on "Supply Chain Frontiers."
- Yeah, thanks for inviting me.
- Can you introduce yourself and why you started the MIT Humanitarian Supply Chain Lab?
- So actually, I came to MIT to help with some educational innovation programs. We set up our first global SCALE center in Spain. That was my first role, to help create a new educational research platform in a different country, similar to what CTL does. And we've had many more of such centers now. When I came back after doing that, I worked on the master's program. I ran the master's program at MIT for a couple years. But all that time, I was looking at what research I wanted to dive into. And I had been doing research with companies like CTL mostly. A lot of our work is driven by industrial applications. But I grew up responding to disasters in Kansas. So I responded to tornadoes. My church and other organizations would be the ones to show up after a tornado had destroyed a house and help them muck out the basement and so forth. And I remember loving the energy of showing up after someone's had their worst day. So that was in me. The first year I was at MIT, there was a student who had worked in the Peace Corps, Tim Russell, and he said, "Hey, I kind of would like to look at humanitarian supply chains." It was not a field at the time, I'll be honest. It was not a study in academia, but I was interested, and we started doing research. Then what happened in December that year, this was 2003, there was the Indian Ocean tsunami, and that was a tsunami of the world realizing, "Hey, logistics and supply chain is pretty important after a disaster." And so there's a whole study that emerged following that big crisis. In fact, the UN organizations, Red Cross, all changed how they approach things based on that crisis. But one of the big things is that logistics and supply chain became more of a profession in this space. And so we kind of rode that wave. And eventually, in 2010, we helped be part of a larger response to the Haiti earthquake on campus, which really then hooked me on, "I need to focus more time on this research." And so in 2011, I started the lab. And actually, Tim Russell, who helped me start thinking about this as a student, has come back to work with us. He's been working with our lab now for over 10 years. So it's kind of a full-circle moment that a student came with an inspiration. I had an interest, and we together started exploring this topic.
- I love the story that you told too of your initial interest in disaster response because it seems like a no-brainer now, where you have a disaster and you need a communal response. You have to have a communal response, and supply chains are the mechanism for that. So your lab works across four areas: supply chain resilience, housing innovation, health emergencies, and humanitarian action. How do these areas connect for you all?
- So the mission statement we have is basically understanding and improving supply chains to meet human needs. So it all starts with a human need. What needs are there? And there are a lot, especially in a post-disaster setting. But in general, the idea of, "How do you have a supply chain react to a crisis?" is the connecting part. Initially, we started by doing aid supply chains. How do you set up a specific new supply chain to bring in aid because the markets are not recovering and not resilient? Especially internationally, that's a common situation because markets are more vulnerable internationally. But even in the US, we have a disaster. It takes a while for the private sector to get going again. And so the initial work was with aid organizations like the UN, the Red Cross, NGOs like International Rescue Committee, Medecins Sans Frontieres, and so forth. These are organizations that set up a specific supply chain to respond. Soon, actually, part of the Indian Ocean tsunami was a realization that if we give people cash, they can go to the market. If the market's ready and resilient, that's a better way to support human needs being met by a supply chain because they know that supply chain, it's in their community, and it has more capacity than setting up a bespoke supply chain. So we started doing research, and this is where Tim actually was part of this initial work with World Food Programme in Sudan on cash and vouchers. How do you stimulate a market, and how do you make sure the market's ready to respond? In our Center for Transportation & Logistics, we study private-sector supply chains. So the resilience of those supply chains, creating a market response, is where most of our research is. And that's where the resilience work comes from. Housing innovation is a specific need where we think there's a lot more opportunity for better resilience. So that's more of a domain, a particular kind of supply chain, construction supply chain to meet housing needs, is a big opportunity, and we're diving into that. Now, health emergencies, health is obviously a sector that has a lot of specialized knowledge, but the supply chains are still structured the same way. So we support health emergencies where there's been a lot of inquiry from the public health profession and medical doctors on how to make the supply chain work better. And then, just in general, humanitarian action, we're ready to support any need, and we just try to find the supply chains that are not working as well in meeting those needs and try to reach out and make those connections, and we always have to be invited in.
- Yeah, absolutely. So thinking about the larger systems at play here, your lab frames problems as systems-of-systems challenges. Why isn't it enough to just solve one part, like building faster or moving fuel quicker? Why do you have to look at the whole thing?
- Not everyone thinks in terms of systems and doesn't realize sometimes how many systems are at play in something that seems pretty straightforward. So you mentioned housing. Most people understand you need a builder to rebuild a house or repair it, but the builder needs cash short-term and long-term, so the banks have to get involved, and then the insurance company has to be involved. In terms of rebuilding, to make sure it's going to be meeting the code for a future disaster, you have architects and engineers involved. The builders rely on a building supply chain, which is its own supply chain. They have a construction process they have to do. And then there's also all the inspections and city planning. So there's a lot of things that are a lot of systems that are required to repair or rebuild a house. And sometimes we think, "Well, it's all just one thing, the builder's doing it," but the builder has to have all those other systems working together to enable them to perform the activity of repairing or rebuilding the house. And so that's the way supply chains work as well, as a system of systems.
- And I'm sure in your line of work, you have a lot of stories to tell over the years. Can you give a concrete example where failure in one part of the system stopped everything? What happened?
- Yeah, I'll tell one that people might remember from 2021, when the Port of LA seemed to be closed. All the ships were queuing up offshore, and they couldn't offload, the big bottleneck in Los Angeles. And you saw in the newspapers people trying to talk about how to unclog that bottleneck. And in fact, the government stepped in to say the port should be open 24 hours to unclog the bottleneck. And they did that, and nothing changed because that wasn't the bottleneck. So understanding the system is required to figure out what actions, what interventions are going to unclog or make the flow work, go faster. And interestingly, this is, and I give a definition of a supply chain, by the way, there's a textbook that said, "All of the parties involved directly or indirectly in meeting a customer need." And so freight transportation is a good example of lots of parties being involved. So you need to have the ocean carrier bringing the ship coming in. You have to have the port being able to offload the container, position it. There's all sorts of equipment in the port that have to position it to get ready for the drayage company. A drayage company grabs the container, takes it out of the port to another facility, like a transloading facility, where it might be put on a chassis to go to a warehouse. So that's already a lot of parties involved in that particular supply chain. So where was the bottleneck in 2021? It wasn't the port. It turned out it was the chassis that enabled the containers to be moved from the port out of the port. And this is something we had actually seen in Puerto Rico following Hurricane Maria, because the hurricane hit, and the port of San Juan was inundated with containers, but the warehouses didn't have electricity and so forth. So when the container arrived, they couldn't really offload it and so forth. So they let it sit as a storage facility, and therefore it tied up the container, the chassis underneath the container. And so the port got backed up because the containers couldn't get unloaded at the facilities, and then the chassis was tied up. So something similar happened in 2021, and one of the easy ways to unclog this was then changing a requirement such that empty containers in Long Beach, they can only be stacked two high for aesthetic reasons. The city of LA, of Long Beach with them, was able to change to stack four or five high so that empties could be brought out and the chassis underneath it could be freed up so it could pull containers out. You needed to know the drayage operators and the other people in the supply chain to be able to figure out that that was the real cause of the bottleneck.
- Yeah, and then, of course, all the communication problems
- Yeah.
- That happened during a disaster too that you have to work around.
- And it actually took, I mean, we had learned this, and thankfully one of the freight forwarders, a big company that does freight forwarding, they did a tweet storm to help the public understand this complex challenge that we see, the system of systems, as researchers, but sometimes the public doesn't. And so that communication of how the system worked, once they did that, he went around in the port and showed videos of all this stuff. Once people could see how a port worked, the system of systems, then you could say, "Oh, that's obvious. That's clogging that. If we free up that, that'll free up that, and then the thing can flow again."
- Pivoting to dive more into housing, which is one of your key focus areas. With climate change, the frequency and severity of natural disasters is increasing. We've had a lot of conversation around housing becoming more unaffordable, in part because of increasing insurance costs due to increasing disasters. Can you walk us through why post-disaster housing is a supply chain problem?
- It's interesting because I think a lot of people think about supply chains as delivering goods or items or commodities, but projects are also a supply chain. So construction is a project, but it's part of a supply chain. There's goods out to come in, but also you're building a house that's innately a manufacturing process. It's a supply chain activity. It's a production activity. And those aren't studied as much in academia for some reason. And because there aren't as many project supply chains in the marketplace, actually humanitarian response is kind of a project supply chain. And in fact, a lot of people are starting to make the connection to, "Why can't we build houses like we build automobiles?" Automobiles are kind of like a project supply chain, but, I mean, it's in a factory, so people don't think about it, but you have to have lots of things come together to create that big product. So it's more complex. So post-disaster housing, one of the challenges is that after the disaster, there's not a lot of resources to meet the surge in demand. There's an ongoing market for housing and obviously, as a lot of us know in the country, there's not been enough supply. When you have a shock, there's even more demand. And so that market has been disrupted. That's why we are looking at post-disaster housing as a supply chain problem, because the local conditions that are required for traditional onsite building are not there to meet the surge in demand.
- One survivor in your research spent 10 months rebuilding a home that took 45 days of actual construction. Can you take us through what's happening in those other eight and a half months, and what does it tell us?
- Yeah, that was interesting. We had a roundtable in May, and I hope we can share the report as part of this podcast. And the story's told in there, but it's interesting because it was actually, the participant was a home builder. So they know the sector. They know how construction works. And hence, the 45 days of construction went pretty well. The other eight and a half months were consumed by engaging insurance companies, the banks with the mortgage, doing the permitting. In post-disaster, there's a need for a lot of permits. And so the local authorities need a surge capacity to process permits and all the coordination steps it takes to get the ability to build. And so that builder had not anticipated that in a post-disaster context, things that work kind of normally in a non-disaster context can take a lot longer because each is a system, each has dependencies on other systems, and they all have to be orchestrated to make sure that you have the ability to start rebuilding and to follow through on the process. Surge demand can really disrupt an efficiently designed system, especially a system of systems. And so when there's a surge in demand, it's not designed to scale up, and that's what we're trying to look at.
- You talk about industrialized construction in the report as a solution, and you talk about also the terminology, how that can be confusing. When we say modular, manufactured, or factory-built, they can get mixed up depending on which audience you're talking to. Can you talk about why this terminology problem matters and some of the issues that it's caused?
- Yeah, you're right. This term, industrialized construction, is still pretty new, and it comes out of trying to address the fact that people most think about manufactured housing as being like mobile homes and trailer homes. And that's a particular way of constructing something that now, due to HUD in the '70s, can be a permanent dwelling, but modular is another way of building something offsite in a factory setting or something and bringing it to a site and living in it. But it's not the same industry as what we call manufactured housing. And so they get conflated, and it's really hard for consumers to know, well, what's a manufactured house? What's a modular house? So part of this is helping consumers know that you can have a house that's built in a manufacturing facility that is not a trailer home. In fact, ironically, some of the highest-end, most sustainable, energy-efficient homes in the country are built using modular approaches or manufacturing. So we've got manufacturing kind of manufactured housing is kind of on the two extremes, more affordable housing and the kind of mobile home space, and then high-end, carbon-neutral houses on the high-end space. And what we need to do is fill in the middle with manufacturing playing a role. There's real implications. So first of all, there's the confusion, right, you know, and so, of these different things for consumers, but also one of the consequences is setting policy with this kind of industrialized, modular, factory-built, system-built space for a while now. And different states have been trying to facilitate that to happen. So Virginia was very early in changing their codes so that modular construction could be a norm across all jurisdictions. It's actually a state law that the local jurisdiction can't set codes that don't allow for this. So different states have different policies, and the way they talk about it is important because the terminology, when you put it into statutory constructs, can make a difference. And I think a lot of times these terms are not well understood, and states are developing laws meaning one thing, but writing something else and conflating the terms. And then there's a lot of work later on to understand what was really meant when that law was written and then interpreted appropriately. And so there's a lot of corrections that have to be done to policies that are developed. And so there's real consequences in not having a proper scientific way to describe this industrialized supply chain of construction.
- Yeah, and part of what's in the report is standardizing, offering clear definitions for each of those.
- The first thing we had at the roundtable is agree at what terms meant because they're used differently in different contexts. And I'll even say there's a massive legislation this summer passed in terms of housing to make it more affordable, the "21st Century ROAD to Housing Act." Even that legislation still suffers from terminology that will have to be cleaned up as we go forward trying to implement it. And fortunately, then, this act actually has within it some calls for research for HUD to sponsor, working with their partner, the National Institute of Building Sciences, NIBS. They were set up in the '70s to help HUD make housing more of a scientific approach. And I think now is the time to think of not only building science, but also management science of supply chains to make housing more affordable by designing the industrial supply chain.
- Yeah, absolutely. And what can industrialized construction do that traditional onsite building can't, especially in a post-disaster context?
- So if we think about to build a house, and actually one reason I'm passionate about this sector is I grew up the son of a general contractor, so I was on job sites throughout my youth helping build houses. And so that typical model is, we ship a bunch of materials to the site, and then you have a general contractor who hires subcontractors who do different roles to assemble the house. And so everything is being brought to the site, but there's inefficiencies in that. And one of the ones that you'll hear most general contractors complain about is, "How can I get my subcontractor there on time to keep the project going?" because time is money, because they have taken out a loan to buy materials, and cash flows are really tight for builders. And so they want to get stuff done faster because they have a loan to buy all the materials to build the house that they're paying interest on. And so time is really important, but it's difficult to orchestrate the production of a house when you have to bring the workers to the site. And you have to bring the materials to the site and assemble it in conditions that aren't ideal. Instead, we could bring the materials to an offsite location, and you don't have to build the whole house, you build a room or even just a part of room, like a bathroom. In a disaster setting, you can do it in a place where there has not been a disaster, where there's not suddenly a demand for plumbers and electricians and carpenters. So the labor market is more stable far away. Second, they can build in an environment that is more conducive to construction. And third, this is one of the big things, is we can think about when we start centralizing and doing things consistently in one place, we can bring in industrialization. And that really means, like industrial engineers, that's my field that I got my PhD in. Once you have a really defined process and a facility, you can really tweak it to make it better. And there's economies of scale and scope that can be realized. And that can make it more affordable. And also, quality control comes into play. You can now do quality control in the facility and not have to rely on an inspector going out on a site and saying, "Hey, did these people who came out one at a time build this the right way?"
- Right, and I imagine there are different rules, of course, for each state, so then there's a lack of standardization there as well.
- And you bring up something there that's really important. And one question is, "If it's so obvious to build offsite and there's a lot of efficiencies, why hasn't it happened?" One reason is because there's not consistent demand. Once you have an industrialized process, you need a more standard product. And if every product is built differently, that's the way our system is built. Each jurisdiction, even local jurisdictions, can say, "This is how you have to build the house. You have to build it this way." And then we send an inspector out to go look at it and make sure you built it right. And that's the quality assurance. Whereas we could flip this around and say, "The code is performance-based. It should just do this." We then inspect the factory to say, "Do the processes in the factory provide the quality necessary to perform as we want it to in our jurisdiction?" And then the quality assurance is done through inspections randomly at different times just to make sure that the process is still delivering the quality that's needed. Now we're not inspecting every house to see if it's built right. We're inspecting houses as they come through the production facility to say, "Is this module built right?" And there's still going to be some onsite work to put things together. And that's one of the questions, is what do we do onsite versus offsite? That's a big research question we're diving into. But when you can have most of the quality assurance done offsite to a performance-based code that could be standardized more, then the whole market can emerge. And right now, it's up to each jurisdiction. Each city or town or county can say, "Build it our way," and the factories can't serve that many different ways of building.
- You emphasize in the report that resilience must be made generally accessible rather than a one-off. What does it take to make resilience the default design?
- Yeah. Kind of building on what we just talked about with the inspections. So, for example, in our roundtable, we focused on how do we not only build faster or cost-effective, really, how do we build better, especially in areas where fire hazard is real. And you could say to people, "Well, don't live next to forests that burn," but that's their home. It's their livelihood. But we can also build homes better, right? And especially if we had a whole community building homes to a higher standard, then there's a risk pooling that happens. If the homes in Altadena now were rebuilt to a resilient standard, then the fire that might come down the mountain in the future is like a firewall for Pasadena. We can build better to then allow ourselves to live in places that have increasing risks, but still have resilience because we've built them better. But to do that, we have to have a standard way of building that is going to provide that resilience. Right now, the ways we build for resilience are through specific program certifications. There's one focuses on wind damage and hail damage for building roofs. It's called a "FORTIFIED Roof." That's a certification that the insurance companies have seen. Now, that makes a difference. It makes the house more resilient, but it requires, again, an inspector to go out and make sure it's built right, and not every homeowner chooses that. If instead we said "FORTIFIED" is now part of a production process in a factory, it can be done, first of all, less expensively because it becomes the default way of building in that factory. And second, the quality assurance is guaranteed so that every house that's rebuilt has it. So if only we had had a factory ready when the LA wildfires happened that could build, and there's an emerging standard called "Wildfire Prepared," if we had a "Wildfire Prepared" factory ready to go and all of the other systems of systems ready to engage that factory, we could rebuild Altadena in a way that is resilient for the future. And that's the vision we're trying to play out.
- This episode is brought to you by the MIT Humanitarian Supply Chain Lab at the MIT Center for Transportation & Logistics. The lab designs supply chains, develops planning approaches, and facilitates cooperation across public, private, and nonprofit sectors to meet the needs of humans in crisis. Learn more at humanitarian.mit.edu. So pivoting to the health emergencies and preparation part of your work, your work directly supports health systems to surge medical supply chains during emergencies. How is surging a medical supply chain different from, say, surging fuel or surging housing?
- Yeah, well, first of all, medical supply chains, there's a different kind of quality standard that's needed for human health, and appropriately so, right? And so that has implications on the supply chain. For example, for treatments that require cold chain or refrigeration, your supply chain has to be designed completely differently to make sure that happens. And also, there's just the quality of the interventions that the supply chain is supporting has to always be there, but also the timeliness, right? In thinking about medical treatment, you need the item to be there when the patient is there and the practitioner's there to administer the treatment. And so medical supply chains are just a special version of making sure the supply chain is having quality and timeliness appropriate to a medical intervention.
- What are some of the examples of the health emergencies that you've been involved in where you've had to apply this work?
- When we're invited in, and any of our humanitarian actions where we're responding and being active part of a response to a disaster, we are invited in. But especially in the medical space, we need medical professionals who need our supply chain and systems engineering. And so one example is in 2014. I have colleagues in Boston who were working with the Liberian Health System to set up after the war, their first reset up there, establish their residency program to train more medical doctors, and then Ebola hit. And so they asked for our support in helping provide some supplies. Then the Ministry of Health said, "Hey, in addition to providing for treatment centers for Ebola, our hospitals need protective supplies so that they can do regular procedures, especially like childbirth, which has a lot of blood exposure, and that's how the pathogen is transferred through blood." And so the doctors were invited by the Ministry of Health to set up a supply chain to support the public hospitals for regular protective equipment for regular procedures, and then they asked us to help set up a supply chain. And that was a particularly challenging time to set up a medical supply chain with that kind of protective equipment because there was such a spike in demand. And there's a whole set of stories on how we had to be creative to respond to that emergency. But that's an example of, with the right medical invitation, our supply chain knowledge can be really useful in setting up an intervention to make sure that the nurses and doctors in Liberia providing regular procedures were also safe while there was an Ebola epidemic.
- Can you talk about some of the preparation work that goes into working with health systems and medical supply chains?
- Yeah, actually, and one of the things, kind of like what I was saying in Liberia, they didn't have the protective equipment. And so one of the ideas is, well, "Why don't we store more of that?" It's a common question. In fact, that came up during the COVID pandemic, is, "Why don't states have more stockpiles to support the hospitals for things like N95s?" I think everyone has some connection with that supply chain. The difficulty of stockpiles are a challenge to maintain because, first of all, you have to have a capital investment in things that you don't necessarily need. And second, they might expire. And so you first have to buy a lot, and then it might go bad, and so you don't want to take the risk. So how do we think about stockpiling for a pandemic or an epidemic? We actually got involved through Mass General Hospital during the pandemic with the state of Massachusetts and the regional states, actually, on how would we set up a stockpile more effectively. And there was a whole study on that that we did to talk about how you could assess what kind of stockpile you might need for a future epidemic that would be similar to, or a variant of, the COVID. Then a few years later, they came back and said, "We love that report. We now have set up a stockpile. Can you help us figure out how to maintain and make sure we have the right items?" And so we just finished a project with the state of Massachusetts on establishing rules for deciding what to buy and put in their warehouse, in their stockpile. And it's interesting because it applies a supply chain concept that is common in academia. We teach it in all of our classes, called the newsvendor problem, but it's difficult sometimes to formulate that problem in a way that you could use it practically to say, "Well, how do I make sure I don't buy too much or too little, given the risks of, if I buy too much, it might expire, or too little, I might not have enough in the crisis?" And so we helped them develop a more practical tool to create the data and kind of create synthetic data to make that newsvendor problem something we could apply directly based on the incentives and the logic that they understood that the health authorities wanted to be able to provide in the state.
- So you would say that now we're more prepared, should another pandemic occur, God forbid.
- That's the idea. And there's actually two things, to be honest, that we've given them because in the first project, we helped them figure out how to quickly assess what demand would be based on a trajectory of a particular epidemic. Second one, this recent one, is, given that, how would you then stock ahead to show that if they had had a stockpile prior to the COVID pandemic, how many weeks they would've had to be able to adjust and find a source of supply for things like N95s. I mean, stockpile's basically buy you time, but that time could be essential because you're going to have to adapt your supply chain.
- Yeah. You all work across many different kinds of disasters, which I imagine makes your work quite challenging. You have to be very adaptable and flexible. What have you learned about preparation that can actually transfer across different types of emergencies, whether it's a hurricane, an earthquake, a pandemic, or other disruption?
- Well, talking about stockpiles buying time, right? It doesn't matter what the hazard is, hurricane, earthquake, pandemic. I mean, they'll have different characteristics. Sometimes you get a lead time. With a hurricane, you have some anticipation. So you could start doing some of the preparation work in kind of real time as you see it coming, right? And that's actually... So in those kind of situations, if you're preparing for hurricanes, you may not need as much of a stockpile. Because you can see it coming, and that adaptation and that getting ready for the hazard can be done in part by just mobilizing really quickly and moving items into position. But for an earthquake, you can't. And so then for earthquakes, you do need a stockpile. It's a lot about understanding the system of systems, which we talked about before. And how can we, regardless of what the hazard is, pandemic, earthquake, hurricane, the issue's going to be that somehow the supply chain and the systems are not able to meet demand effectively because of demand surges and because supply chain capabilities are disrupted. Warehouses might be out, roads might be severed, and so forth. So we more look at the types of ways that the disruption hazard manifests in the system, and then we prepare that way. So we're really preparing across all kinds of hazard types by just understanding demand and supply systems in a better way.
- Looking more largely at your humanitarian and global work, as you said, you've worked on several continents with development and community organizations in direct action. How does working globally change what you learn about resilience?
- Yeah, we have had the privilege of working on several continents, and sometimes it's interesting to realize the country I know best is the US. I know the government and the industry settings best, but sometimes working out of that context helps you see things differently. And we're trying to do something that we learned from the international sector, from the years of working internationally in disaster response, from something called the Logistics Cluster. So this is part of a humanitarian orchestration system that emerged out of that tsunami in 2004. That tsunami response is called the cluster system. So different medical specialty, different specialties like medical, water and sanitation, food, they have sectoral focus on how to meet people's needs, but there's also a Logistics Cluster on how to move all the items to support the food needs and the water needs and so forth. And so that Logistics Cluster is something, is an organization we've worked with for years, done a lot of research with them. And some of their approaches to how to orchestrate the supply chains that are trying to respond after a crisis are things that have not been fully adopted in the US by the national disaster management, the FEMA, or state disaster management organizations. And partly because they haven't been exposed to them, but in part because a lot of the supply chains that are needed to respond are private sector. And so there's just sometimes a difficulty in being creative between public and private sector to solve a problem. And that's where academia can play a role. And so we're trying to bring some of those concepts on how internationally they mobilize logistics resources for the common good in a community and apply them to communities in the US with the local government and the local retailers and health system providers who are all working together within a community to do something similar.
- So looking across disasters, you mentioned in a December 2024 roundtable on fuel supply chains that fuel is the number one concern in disasters because it enables everything else to happen. Can you speak more to that and why it matters so much?
- In a lot of humanitarian and emergency management organizations, they don't really explicitly plan for fuel. And it's always surprised me because in a post-disaster setting, fuel enables you to move things, trucks and ships and everything, but also you often lose power. And so electricity relies on generators that rely on fuel. So it is the essential commodity for everything else to work. I think it's because it's not something you often give directly to people who have suffered. Obviously, they'll be seeking fuel for their own needs, but they kind of miss out on the fact that if we don't have fuel, then we don't have the ability to respond. We got really involved with FEMA in enabling the private sector during their 2017 hurricane season. And when given the opportunity, they sponsored a National Academies study to see, "How do we think about supply chain resilience?" We decided at MIT to focus on fuel because that enables all the other supply chains. And we have learned a lot about how fuel supply chains work in the US over the years and kept doing work with FEMA on understanding how the upstream supply chain of extracting oil, refining it, getting it to a terminal, is often very separate and decoupled from the delivery to a fuel station to make it available to the public and the emergency responders. And so we've tried to close that loop in understanding that delivery, that last-mile part of the fuel supply chain, in a much better way to make it more resilient to surges in demand because you're going to get a surge because, again, fuel is the basis for everything in a response.
- I remember in the workshop, one of the participants mentioning that one of the big variables in disaster responses that you can't control for is human behavior. I remember him saying when people would see the fuel trucks coming in somewhere, following it with guns and waiting to be first in line, for example. How do you prepare for things like human behavior in a disaster context?
- Well, you can't fully, right? Part of it is experience and having seen it before, thinking through, for example, the thing you mentioned is that a lot of times they have the provision for escorts to bring fuel trucks to stations for two reasons. One is get some access to the station because if there's a line of people trying to get fuel, it's hard to get the truck in. So they make sure that there's a way to get the truck into the station. And secondly, if the truck arrives, that the surge for demand does not cause behavior to disrupt the fair allocation of that fuel to societies. But the bigger thing is, one thing you can't control, for example, in fuel, we did a big study with FEMA following, in 2021, there was a ransomware attack on the Colonial Pipeline, which distributes most of the fuel on the Eastern Coast, long story short. And so news got out, and so everybody, human behavior, panicked and went to the station to go buy fuel. So it turns out during that whole Colonial Pipeline ransomware attack, the terminals that had fuel near to the stations never ran out. But there were shortages and long lines at stations because people, again, behavior, went to go buy more than they needed. So if the word hadn't gotten out, no one would've noticed that disruption. In fact, there had been disruptions to the Colonial Pipeline of that duration in the past that did get publicity, and there was no issue in the station. So part of it is, you have to anticipate that in a disaster, human behavior will kick in to make demand really surge. And that's where we have to think about supply chains that might be able to respond to that human-induced demand surge that makes the supply system fail.
- We saw a lot of that in the pandemic as well, with the empty shelves. Looking ahead for supply chain professionals listening, what's one thing you wish they understood about building resilient systems that might change how they think about their work?
- We did a roundtable actually in the spring of this year, in 2026, on supply chain resilience and why it's talked about a lot now, but how do you actually get a company to invest in it? People say, "Well, we're investing in something we may not need." And how do we justify that to our shareholders if you're a company, for example, or to the public? If you're a government agency, you spent money on something that you didn't need, so. And what I often say in response is, if you're trying to understand resilience, but how you would make your system react and respond to disruption, you're learning about your supply chain. You have to know more about your supply chain to figure out how to make it adapt when disruption happens. When you learn about your supply chain in that way, you'll find a lot of spillover effects because you now know how your supply chain works better because you've tried to stress-test it. And by stress-testing, you can find opportunities for innovation. And I think that a lot of companies don't realize that investing in resilience and understanding your supply chain better so it can adapt better when needed gives you opportunities to reinvent how your supply chain works day-to-day. And that spillover effect is, I think, vastly underestimated by a lot of companies. We would love to work with companies who want to take a chance on investing in resilience and seeing what kind of spillover effects happen, because I guarantee that there's going to be a lot of day-to-day benefits come from understanding your supply chain better.
- Yeah. It seems like all of us can count on disruptions of one form or another.
- In fact, at that roundtable, one thing I talked about is another way to position investment in resilience is like buying insurance. We buy insurance we hope we never have to use. And so we also played with that framing internally in the company, is, if you know your supply chain really well and you can adapt better and faster to meet customer needs, that's the kind of insurance that your CEO wants to make sure that the quarter-to-quarter financial results are going to be good. And so instead of seeing it as an investment we might never need, it's insurance to make sure that our supply chain always performs like it needs to.
- And just one last question to close: what gives you hope about solving these problems?
- We're leaning much more into more direct engagement with communities to help shape the supply chains that serve them. So in our work with housing, I think a lot of this is, how do we help local leaders and local developers and local builders and local insurance agents and local banks all say, "Hey, if we all together said we want to build houses differently," then the supply chain might be set up differently to meet our housing needs, right? And I think if that kind of community action to mobilize demand for an innovative supply system can happen, then I think we will get better results in society. And so I think that can happen with housing. We're also hoping to do this with community leaders, like working with food banks and grocery stores to say, "Hey, how do we feed people after disaster?" If we talk together, maybe we can collectively have a better supply chain to respond to meet people's needs if there's a need for that. So I do believe that community-led design of supply chains is something that can have a material impact on disaster response. And that preparedness, again, it's hard to sometimes think about preparing for something that you hope doesn't happen. But I do believe even if communities say, "How do we feed ourselves in a crisis?" They'll think better about how to feed themselves day-to-day and meet human nutrition needs on a regular basis.
- That wraps up this episode of "Supply Chain Frontiers." A big thank you to Jarrod Goentzel for sharing his expertise and insights. "Supply Chain Frontiers" is recorded on the MIT campus in Cambridge, Massachusetts. Our sound editors are Dave Lishansky and Danielle Simpson at David Benjamin Sound. And our audio engineer today is Kurt Schneider at MIT Audiovisual Services. Be sure to check out previous episodes of "Supply Chain Frontiers" at ctl.mit.edu/podcast or search for us on your preferred podcast platform. I'm Mackenzie Berry. Thanks for listening, and we'll catch you next time on "Supply Chain Frontiers."
File
