 Hello everyone, welcome to this webinar focused on strengthening supply chain resilience. My name is Jared Gensel, I'm at the MIT Center for Transportation and Logistics. This time I'm going to hand over to Stephen Stickter from the National Academies. Greetings, Stephen Stickter, Director of the Resilient America Program at the National Academies of Science, Engineering, and Medicine. Thanks for joining us. I'm going to give you a brief introduction to the study itself, how it came together and handed off to two of the committee members who led the study and led to the findings that we're going to be presenting today. So just as a background, the study focused on observations from three events in the 2017 hurricane season that caused significant damages, disruptions, and deaths in the southeastern U.S. and across the Caribbean. Next slide. Three major storms, and Harvey, Irma, and Maria, and the storms happened in very close succession. The impacts that we're talking about began in, that we'll be studying in this, in this study, began with the landfall in Texas of Hurricane Harvey on the 26th of August 2017, continued through Irma in the Caribbean, and then it's, travels through Florida and followed by Maria, and these storms in the impacts that we're looking at all happened within a one month period. There are some significant reasons that these, these storms led to this study or led to interest in, in studying the impact of supply chains in that there were multiple areas that were affected in close proximity to each other in the same region of the U.S. and some of those areas had multiple hits across multiple storms. We've had great advances in the last number of decades in forecast and forecasting for tropical cyclones and hurricanes, but forecast uncertainty remains. And combining that with rapid intensification of some of, a number of these storms proved really challenging for emergency preparedness and evacuation recommendations. And these events cover both wind damage events and also significant flooding, particularly in the case of, of Harvey in Texas. Our focus today will be on the supply chain impacts and, and the observations and recommendations that came out of the study committee on how, on how we can make those supply chains more resilience, resilient to future impacts. Next slide. The committee's task and this, this study was carried out by a committee gathered by the national academies and the, and the committee's task was to use the observations and experiences of the 2017 hurricane season to understand the impacts on supply chains and for that analysis needed to understand and gather information on the strengths and vulnerability of supply chains to produce lessons learned for the, for future implementation. And in this work, the, the supply chains that we're looking at are both the public sector and private sector and very importantly that, that interface between those public and private sector supply chains. Next slide. This was a three part process that the overseen by the national academies, the first phase of the project was a deep dive by CNA to collect data information, interviews on the ground experiences of people and places that were affected by these storms to gather preliminary insights and feed into the further analysis and work by, by the national academies committee. CNA's work had on the ground components in Florida, Texas and Puerto Rico. Then the national academies committee came together and using information from the CNA plus on the ground visits that they took in the regions affected by these storms with the purpose of producing findings, conclusions and recommendations for further work regarding supply chain resilience. And this included both public meetings in throughout the, throughout the report writing and information gathering phase. Finally, we worked with the MIT transportation and logistics lab to produce analysis of specific supply chain components to better understand some of the blockages and connections between, between supply chain components to inform how both disruptions happen and ripple throughout the supply chain and also to provide some concrete ways of implementing some of the, the recommendations included in the, in the national academies report. Next slide. CNA as noted was the first on, first on the ground in this, in this project and did a series of deep dives both into the human and the market impacts of the storms and supply chain from, from those storms and included a wide range of supply chain components and, and also write down to what are the impacts of donations on, on response and recovery. Next slide. The national academies committee encompassed a broad range of experiences from emergency management, academia, nonprofit, private sector and practitioners and government today on the, on this webinar, we have Oslam Ergun from Northeastern University and Craig Phillips, Philip from Vanderbilt University who will be like guiding you through some of the findings and then recommendations that came out of the study. Next slide. And throughout the, throughout the work of the committee, committee traveled to and held public meetings and, and meetings and private meetings with officials and, and impacted residents and businesses in Washington, D.C., Puerto Rico, Houston, Miami and St. Thomas and the Virgin Islands. So I'll pass it on to Craig. Thank you very much, Stephen. And good morning, everyone. Thank you for that introduction and Jared and the MIT team. Thanks, thank you for setting this up. I know we're all disappointed that we couldn't be undertaking this in person, but we have a new normal for a while. My name is Craig Phillip. I'm a faculty member at Vanderbilt University. I lead our transportation and operational resilience center. But my background is primarily as a, as a, as an operator and practitioner. I spent 30 years in, in the maritime sector as an owner and operator of boats, barges and ships operating primarily domestically in the United States. We did have during my time there, we had active operations in Texas and in Florida, though we never operated in the trades to and from Puerto Rico. I'm going to, I'm going to kind of cover some basic supply chain concepts that, that the committee spent a little bit of time at the beginning trying to, trying to focus on the, on this very large domain. There are of course, multiple frameworks that have been developed over the decades to understand and analyze supply chains. And we had several members of our committee that were, that were well regarded as being, as being creators of some of those frameworks. But we recognize that we needed to simplify the way that we looked at this domain, make sure that it fit with the, the context of looking at, at disaster and disruptive events. And so I'll go through those very quickly and then I'll cover a little bit of our kind of macro findings and then turn it over to Oslo to, to, to dive into the, some of the recommendations that came out of our work. We decided to settle on two principal concepts that, that kind of, that kind of can characterize the way supply chains are operating. The first is the concept of a bottleneck and the, the notion that there are spots in any supply chain where the, that are going to determine the overall capacity of that supply chain. And the second is lead time that, that the supply chain is in part characterized by the length of time it takes for products to be delivered once they're ordered. And in disasters, especially that, that issue can be particularly prominent. We, we looked at, we decided to focus our, our, to characterize the types of disruptions to the supply chain into three different buckets or categories. The first is, I'll call demand shift. And as you'll see when I go over some of the results, there was a significant spike in the demand for fuel associated with one of the, one of the events. All of us today, I think, are, are facing the, the, the demand shifts that come from hoarding, which seems to be prominent in the, in the response that, that we citizens have, have characterized us in this, in this current environment. The second and maybe the most pronounced disruption is, is reductions in the capacity of the supply chain itself that can start at the, at the origin of the, the production of a, of a good to be moved or can go into any of the elements of the supply chain. And finally, we recognize that the important and prominent issue would be the, the communication supplies, the chain, if you will, that, that accompanies the movement of the physical goods. If you will, the movement of the bits and bytes that necessarily accompany the movement of the, of the physical goods. Next, next slide, please. In terms of the different types of strategies, which we were confident we would, we would be able to, to explore and both their, both the, the successes and the failures, we decided to divide our, divide the strategy environment into, into, into three buckets. The first being readiness or the efforts that are undertaken before a disruption occurs. The second being the response itself, the, the, the, the short term activities that, that both public and private sector actors undertake to replace the, the brakes in the supply chain, which are usually a company, a, a disaster environment. And finally, the recovery. And in particular here, we mean the recovery of the normal supply chains so that they can, so that they can begin to function in their kind of blue sky state and how long that takes. You know, finding the right balance between these strategies and is, is often related to the, to the assessment that's been made, both pre-event and during and after an event in terms of the criticality and vulnerability of these different links and nodes that are in the supply chain. The, so we looked at, we looked at the disruption through this additional lens as well, trying to characterize where the, where the disruption occurred and understanding this, this linkage between criticality and, and vulnerability. Next slide please. Of course, geography is, is of critical importance. And we were presented with, with three rather starkly different environments where these, where these hurricanes impacted the, the United States and its territories. And so we spent some time thinking about that and trying to characterize the, the different supply chains that were, we were going to be, we were going to be confronted with. Starting with Houston, the, it's of course along a coastline, but it's, but it effectively has access 360 degrees. The, there is tremendous redundancy in the, in the, in the supply chain system that exists in the Houston metropolitan area. There is lots of modal diversity, well-developed rail system, a very well-developed highway network, and one of the richest port environments that exists in the United States. And beyond that, the, the Houston, the Houston port complex itself is one of the most important links in the, or nodes in the, in the, in the national and effectively the international supply chain, especially for petroleum and petrochemical products, products worldwide. Turning to Florida, really a starkly different characteristic in terms of the geography at least, a very long peninsula, arguably the longest in the United States. It's got a couple of concentrated high, high density corridors, but it's a highway-centric environment. While there is some rail, the, in the, in the state, it's, it's, it's certainly limited character, relative to what you would find in the Houston area, for example. Ports also play a role, of course, but there isn't a single point of concentration as there is in, in the, in the Texas area. Finally, the Puerto Rico and U.S. Virgin Islands, a starkly different complexion, a very long supply chain from the primary, from the primary source of goods coming from the domestically from the United States in Puerto Rico itself. A single point port facility at San Juan that handles the vast majority of the cargoes that, that move to and from Puerto Rico. And this same port is the single, it's not shown on the map, but is the, is almost the exclusive gateway for products that move to and from the U.S. Virgin Islands. On the, on the island of Puerto Rico itself, a single mode highway dependent system. And as I mentioned, the, the port of San Juan is a, is a, is really the single point of access and egress to the, to the, to the port, to Puerto Rico itself and to the, on to the U.S. Virgin Islands. Next slide, please. The, I'll, I'll, I'll kind of give a, one example of kind of the dependencies that is, that we found that certainly exists between these temporal and geographic features. Puerto Rico found itself at the very end of a long and thin supply chain that, that, that quickly went beyond the breaking point. And this extended to the, to the, to the islands over in the U.S. Virgin Islands as well. And so disruptions that occurred both on the trucks, truck delivery side within the, within the domain of Puerto Rico itself and the, and the, the maritime, the marine carrier operations, which, which were dependent upon almost exclusively for the movement of goods into the disrupted area, the disruptions really cascaded throughout the supply chain back into the heartland of the United States as well. One of the things this, this presented to the emergency managers was the, the need to, the need to make choices between affected locations as they face these, these critical limitations on the capacity of the system and the bottlenecks that presented themselves. And just a graphic, a perfect illustration of how these, these inter interdependent supply chains, especially presented with the unique features of the geographical territory that was being faced could, could collide and cascade with one another. Next slide please. So let me turn briefly to, to talk about some of the high level impacts which we found in the, in the, in the four areas. The, the Hurricane Harvey in Houston was a wind event to some degree, but the, the overwhelming impact was the record, record shattering floods, which inundated the much of the, much of the metropolitan areas. And the flooded facilities that, that were presented to the, to, to the populace post really during and post because it was not only record shattering in terms of quantity of flooding, but it was record shattering in terms of the duration of the, of the, of the, of the rainfall event itself. And there were significant facilities across the region, which were inundated. But I think our, our, our, our short review of the, of the responses that the, the system in, in the Houston metropolitan area really benefited from a level of disaster preparedness and response, which, which I think could be the envy of maybe, maybe the world. And what was particularly striking was the degree to which the public and private sector preparedness activities came, came, came together and played, and the, and the area certainly benefited from the fact that this concentration of petrochemical facilities, each of whom had their own very well-developed preparation and response protocols. And, and there's a, there's good practice in, in, in the, in Houston and Texas for these groups working collaboratively with the, with the public sector entities that are activated during a disruption event. Next slide, please. Turning now to Florida in Hurricane Irma. Florida expects hurricanes there. They have a well-developed preparedness and response infrastructure. They would, they would say they're good at it. And our, our investigation would suggest that that's, that that's true. Certainly compared to earlier hurricanes, the hardening and upgrading of the power systems in, in Florida throughout the state were, were particularly prominent in terms of the benefit that that, anew word to the, the recovery and response and recovery activities. So the, the most, the most significant vulnerability that, that, that was exposed during these events in our, in our assessment was created by the massive evacuation that was ordered anticipating the inter, anticipating Hurricane Irma. Many manifestations of that evacuation, but one of the most prominent was the, that the fuel demand from the evacuation activity itself was unprecedented and the system was not equipped to rapidly recover and resupply the, the, the, the supply chain pipeline for, for, for diesel and, and gasoline and not on the slide, but it also much of the fuel of the refined petroleum products that arrive in, in Florida arrive at two of the four main port facilities, both of whom experienced significant bottlenecks in the, in being able to, in being able to serve the, the truckers that were largely the, the, the final mile delivery of the, of the, of the refined petroleum products. The US Virgin Islands were, were, were struck twice as was, was Puerto Rico to some degree, but the Virgin Islands in particular faced tragedy associated with both Irma and Maria, the, all of the manifestations of the, of the single pipeline, that supply chain pipeline coming through Puerto Rico were, were manifests there. But probably the most significant for particular feature that, that struck the, the, the committee was the, the difficulty associated with disposing of the storm debris and having, having experienced over the last two weeks, the very large tornadic impacts in middle Tennessee while we've had issues here in, in, in collecting the storm debris, the, the, the disposal has not been a problem because we have so much territory to, to move to, but in a, in a, in an ocean, in a landlocked, small island environment, the, the process of disposing of the storm degree itself is a huge supply chain issue, which was not particularly noted or, or in the plans that, that, that have been made for the, the islands. And if we can move now to the next slide, please. Turning now to Puerto Rico, which was hit by both, by both Irma and Maria as well, but, but most significantly by Maria. I'd say that the second bullet item on the fragile aging power and communications infrastructure was the, the most prominent feature, which, which made the, both the recovery response and recovery activities problematic for those, those who are struggling to do so. These lifeline systems were, were weeks and months in, in the recovery mode themselves. And so the bits and bytes that need to accompany the, the movement of goods simply, simply was not possible. The second feature, which, which became so apparent was that the, the, the focus on a single port of entry in San Juan, while the port facilities themselves are well developed and certainly capable of handling volumes that in a blue sky mode, and even somewhat larger, the port of San Juan itself became an extraordinary bottleneck, a variety of physical and institutional factors contributed to that. But the, the, the dependency on that one point of entry became a crucial problem. And then finally, the, there were manufacturing facilities in Puerto Rico, which, which were critically important to the, to the, to the United States marketplace itself and the variety of medical supplies that, that were produced almost, almost exclusively in Puerto Rico and, and where these facilities were impaired and there was not a sufficient ability for the decision makers on the ground to, to have visibility about the importance of these manufacturing plants as they were prioritizing and doing their, and doing their work. In general, the, the disruptions that occur in the supply chain seem to be at the distribution level. It, the, the retail end was, was generally speaking, it appeared to us to be quicker to, to, to, to recover than the, than the distribution channels to get to those, to get to those, to get to those retailers and in Puerto Rico in particular, the, the lack of phone and telecommunications of course made it difficult for these retail entities to, to get back in service even when they, even when they were resupplied. It was certainly apparent that the, that large companies, and this was across the, the four territories have capacity and well-developed in many cases continuity planning and partnerships with government officials and I guess it's no surprise that smaller, smaller businesses and enterprises are less likely to, to have, have these relationships developed in advance. Can't overstate the importance of the critical infrastructure, back bone systems, power and, and, and, and, and communications and to some degree in Houston water supply and the, how difficult it is to, to recover in any aspect that supply chain included if these critical infrastructure systems are, are, are not back in operation. And then the, the fourth bullet is a, we'll hear a little bit more about that later from Oslo, but the, the, the response activity you could call a push activity of, of, of, in a, in a time of, in a time of short-term need pushing critical elements like food and water into a, into an impacted area. But the emergency managers that are responsible for that, that temporary pipeline often don't have good visibility about the, about the status of the, of the, of the, of the, of the recovering private sector supply chains. And so we found evidence that the, that there were, there could be cases where that, that push of recovery materials was overwhelming the, the, the speed with which the, the, the private sector supply chain was getting itself back in service and could it in some cases be impeding it. And I think with that I'll, I think I'll, I think I'll wrap up and, and turn it over to Oslo. Thank you Craig. And next slide please. Thank you. Hi, this is Oslo. I'm a, a Stephen introduced me. I'm a professor of industrial engineering at Northeastern University. And over the last, a dozen years I have focused in looking at humanitarian response, emergency response from operations and systems point of view. So what I'll do today is I'll cover the four high level recommendations that our committee came to distill out of all our interactions over the lifetime of the committee and with all our meetings going over the CNA study and interacting with MIT as well. And just to give a structure to, to our recommendations, the way that we see it is that recommendation one is the key recommendation. And recommendations two, three and four are kind of things that needs to be done, that need to be done to support recommendation one. So, so I'll start with recommendation one, which as I said, which is the high level, high level, the big recommendation, which is to shift the focus from pushing relief supplies to ensuring that regular supply chains, regular mostly commercial supply chains are restored as rapidly as possible through strategic intervention. So while everyone agrees that relief supply chains, which is kind of what Craig was saying as like pushing relief supplies to the disaster areas, surely can be lifesaving following a disaster. At the same time, we have to, you know, everybody has to kind of get to an understanding that pouring these relief supplies into an area for too long can have unintended effects. And we have heard over and over of examples around this theme. And this can happen because the relief supply chains often contract or deal with the same resources that if, you know, if a relief supply chain goes in and take over all the truck capacity, let's say, that would be a resource that would be taken away from the local businesses who are also trying to move their goods around. The second issue that we have kind of identified as a problem with pushing relief supply chains, pushing goods through the relief supply chains for too long would be the fact that this approach doesn't really address the problem of last mile distribution problems, which are usually a consequence of the impacted critical infrastructure such as block roads, lack of fuel supplies, power outages, and worker shortages. On the other hand, we understand and, you know, we do understand the fact that this is a hard dynamic to change that is going from the pushing the relief supplies to ensuring that the regular supply chains actually work because FEMA, as most other relief agencies, is actually judged by the fact that how much relief supplies are pushed. And also a lot of the elected officials and the local communities ask for these relief supplies over, you know, in usually in amounts that might not really reflect what is needed or what is the most crucial role for FEMA to play. And the second thing is most places lack the systems to gauge the conditions and the capabilities of the local supply chains. So there is this visibility problem in terms of understanding when to scale back the emergency response and how to maybe scale it back in categories. So next slide please. However, you know, like we've been talking about what are the strategies for augmenting this focus from delivering relief supplies to actually a focus on finding causes of the unmet demand. Like why is supply, why is demand, is net, is not met with the supply? And again, going back to what Craig has described in his supply chain primary, we really think that the focus should go to identifying the supply chain bottlenecks and the gaps and the broken links in terms of the infrastructure. So some of the key strategies that we have, you know, come up with or highlight from existing literature is to improve the pre-disaster planning and communications, which will help with understanding what are the critical points, what are the vulnerabilities, and what are the capacities and the limitations of the key stakeholders. This is something that came to, this is something that we discussed over and over again during the meetings where there was clearly a gap between what FEMA or the other federal agencies actually knew even before a disaster happened in terms of, let's say, the footprint of a certain critical supply chain in an area that is vulnerable. So the second strategy that we talk about is the expanding the scope of, potentially expanding the scope of FEMA's role from, you know, really pushing the aid to actually more actively aiding infrastructure repair. And that will give a latitude to prioritize support for critical private sector facilities and could potentially latitude to prioritize support for critical private sector facilities. And, you know, the best example for that is what Craig has described in Puerto Rico with what happened with the pharmaceutical manufacturers who are actually almost exclusively producing some of the, you know, medical products and pharmaceutical products for the entire nation. And then also, again, coming back to the political pressure of the local agencies and potentially the public asking for relief supplies. However, that not being the best thing maybe to be done. Again, changing how FEMA is evaluated, not just the, not by just the amount of relief supplies delivered, but also the speed and the strength of the recovery, which really depends on the recovery of the critical infrastructure. Thank you. Next slide, please. So that was the high level big recommendation. So the second recommendation that is in support of the first one is to build system level understanding of supply chain dynamics as a foundation for effective decision support. So the first big issue that we would like to emphasize is that supply chains are systems and an impact. So to understand the critical and vulnerable nodes and links in a supply chain actually really requires a system wide understanding of the supply chain and the dependencies and the cascade that dependencies within the supply chain and the potential cascading impacts that might come at different geographical regions or at different times. So the temporal and the geographical cascading impacts are very important. So we really think that we have to understand the system dynamics both before the event under blue sky conditions, but also after the event in gray skies. So before the event, we have to understand the supply and demand dynamics, the flows and potentially have certain disruptions that might be probable disruptions in a given area might affect these flows as well as the and this will lead into an understanding of the criticality and the vulnerability of key supply chain nodes, links and infrastructure. However, this is not enough because when a disruption actually occurs, the critical and vulnerable points of the supply chain might actually move. And you know, like a bottleneck in a blue sky might actually not even present itself as a bottleneck because a big disruption on another part of the net supply chain network or the infrastructure that supports that supply chain network might actually become the bottleneck. So it's also important to understand the real-time impacts and get the real-time information to understand these impacts on things like the block rows, closed stores or factories, you know, power requirements and so on and so forth. And then we also have to understand the current capacity of local stakeholders to respond to these impacts such as their inventory levels, supplies, how things are within their supply chains so that we can so that FEMA can actually effectively help to bring them back to normalcy. Next slide please. So one of our recommendations is that actually there is a need for modeling and analytical frameworks to do the system-wide understanding of the vulnerabilities and criticalities in supply, in important supply chains. And these need to be then integrated into decision support information to help FEMA to make important decisions in short amount of time where the resources and the time is very limited to actually maybe do huge analysis of what's going on at the moment. So obviously there are benefits of investing in such systems. One is the understanding of the vulnerabilities and the gaps. The second is these systems enable the emergency managers to make effective prioritization decisions time and time and time again I keep quoting one of these interviews that we have had in Puerto Rico where we talked to an emergency manager where he actually said that one of the pharmaceutical companies asked for a generator or fuel for generator. I don't remember exactly which one this was but this emergency manager with all good intentions tried to figure out if this was a critical need. And he actually couldn't figure it out like nobody could answer him and at the end he decided not to give this ask. And lo and behold six months afterwards we still had national shortage of sterile fluids in the country and having even heart surgeries being canceled due to the fact that the manufacturing pharmaceutical manufacturing capacity for all sterile fluids in Puerto Rico was basically did not recover for six months after Maria. So but but at the same time you know if we can actually provide these frameworks understand these vulnerabilities you know the footpath or the footprints of critical supply chains prior to a disaster happening we will enable emergency managers to also make better decisions in a short amount of time. And of course when we have these kind of decision support systems you know strengthened by modeling and analytical frameworks then we enhance the capacity of the emergency managers to focus on strategic restoration of broken supply chain links and infrastructures. Next slide please. So moving on to the next recommendation so how do we actually enable this even this analysis right how do we actually build a system level understanding both for for supply chains both in the blue skies and the gray skies. And one of the key things for this to happen actually we believe that there is a need for support mechanisms for coordination information sharing and preparedness among supply chain stakeholders. So you know some examples include the developing and regularly updating emergency preparedness and continuity of operations plans continuous conduct conducting continuous training and worst case scenario drills test emergency communication protocols and and make plans for what happens if these protocols actually fail and also make plans to protect organizational personnel during host protect organizational key personnel during hazardous hazardous events. So I just see I just saw a question from the first question on the question and answer session and there was a question about earlier when Craig was talking about why did Houston actually develop such a good preparedness you know why were why was Houston so prepared compared to the other regions and this is not a complete answer but one ends I think a partial answer you know presented itself during our interviews in Houston and I particularly remember that especially when we were talking to Port of Houston which is a such a key you know focal point of supply chain and transportation there was the fact that Port of Houston had a lot of small events happening throughout the year all the time so all of their emergency protocols and activities are actually tested all the time throughout these through these small events so the people different stakeholders knew each other the the protocols and the systems were tested throughout the year so when something really big happened nobody was kind of looking at the book and trying to figure out what was the next thing that they had to do they knew the people they knew how the people behaved they knew the different stakeholders and their preferences so it sounded like one of the reasons at least was why Houston actually managed the managed Harvey in in a successful way was because they were they were practiced basically they were practiced on what they were doing Oslo if I can just add to that I know we're not supposed to answer questions in the middle but we'll go forward anyway Jared yeah in Houston I was going to add that the the the entities that are involved in moving hazardous cargoes are required to do drills and as part of the as part of the regulatory requirements of having the permissions to operate those hazardous materials so that's another it goes along lines of what you said they practice a lot both because they have small events but also because the the regulations that govern the operators in that setting have a have a legal requirement to do so thank you it's perfectly good to jump in there Craig I was going to ask you to talk about that specifically because I remember observing the same thing in Houston so yeah all right thank you both and then given that now I'm actually responding to questions I saw the second question about the increased visibility into the conditions and capabilities of supply chain during a disaster I would say I mean that's really the response you know the response to that question is to build these mechanisms and information sharing protocols beforehand and have a way of and have an understanding of how to actually do the communications so you know critical need as I said in the slide is to define the processes for coordination and information sharing across all the all the stakeholders including private sector silos so because one of the things that we have also seen was there are definitely private sector silos where the the government and some of the other partners in the supply chain really have no information about these silos in their conditions or what they might need their footprints what it might mean if there's a problem here what it might be in mean to the entire sector in in a month in two months and so on and so far but we also understand that establishing these relationships and the trust takes time and it definitely has to start before disasters we also understand the fact that there are sometimes even legal barriers to these kind of communications another story that we have heard in houston was the fact that with some of the energy companies fuel companies they actually fema couldn't talk to these companies in one room due to antitrust laws and so on and so forth so we do understand that there are sometimes some limitations but having an understanding of all of those issues and then creating a communication protocol given the constraints is essential before something happens especially for areas that are vulnerable and under constant threat of disruptions next slide please and we also recognize the fact that there are existing mechanisms for emergency response officials to interact with the key stakeholders and in in many circumstances these actually do work really well such as the the dhs critical infrastructure threat information sharing framework and some of the others that are listed on the slide one important aspect that we also kept hearing and we wanted to bring forward is the fact that this engagement ecosystem this interaction should also be done in a way that minimizes the burden on the individual the time burden on the individual participants because we all understand that in the in the eye of the disaster everybody is actually very busy doing what they need to do to respond and we don't also want to create systems that are really burdensome on individuals who are trying to do another job at the same time and again i want to emphasize the fact that if there are systems and there are decision support mechanisms and some information that was created before a disaster happens it will all of these activities will minimize the burden the time burden when actually a response operation is happening the next slide please so coming back to what are some of the effective examples of strategies for increasing coordination and information sharing among the different stakeholders include and i'm not going to read through all of these but again identifying and strengthening industry government coordination for prepositioning and distributing resources i see another slide about you know given the lead time should we start pushing as soon as we know something has happened before doing a need assessment again the question i would say probably the answer is it depends it depends what's actually what is the current situation well it depends in the sense that what is the supply chain the commercial supply chains are like in the in the in an area and again taking taking understanding what the current situation is might take some time but if there is a constant coordination and communication maybe that time could be minimized so another key factor is to put supply contracts in place ahead of time for very key items such as fuel and generators again which will definitely decrease the lead time to react so that you actually have more time to figure out what's going on and then putting the system into gear one important part issue that came out came up came up all the time was the need for the workforce and the need to take care of them and address their needs and their families needs during a disaster so that the workforce is actually available to be part of the supply chain and the functioning of the supply chain so there and given the current situation there are also a lot of reports about the personnel and the potential of supply chains being broken due to workforce shortages and this is a very important issue I think that needs to be considered and then strengthening of mutual aid agreements within lifeline sectors is another issue that came up over and over again again kind of summarizing or putting together a couple of these bullet points is the fact that the partnerships needs to be established before something happens small some exercises and events that will actually bring together together the stakeholders and test the system seem to be key we should also remember the fact that there are a lot of NGOs and other community organizations in disaster areas and there needs to be they need to be considered as partners and maybe included within these information sharing protocols and some of these exercises and again like coming back to some of the other things that I've already talked about collecting data understanding the system that's being in place for important critical supplies such as drugs are our key and I think I'm gonna go to the next slide and finally you know having said all of those recommendations one two three we also believe that there is a need for actually developing and providing training on supply chain dynamics and best practices for private and public partnerships so we realize that a lot of the emergency sponsors have very little or no private sector experience or they're not really trained to evaluate supply chain impacts of disasters and and how to respond to them this so we propose that emergency managers and those that are in supporting roles should be you know should have some basic overview supply chain training and understanding of how critical infrastructure supports these supply chains and some of the laws and regulations that are governing disaster operations and how they have these laws and regulations are impacting supply chain operations and you know there are several recommendations within the within the committee's report about how these training modules can be created or delivered to the emergency responders thank you next slide please so just wrapping up that's the end of the recommendations so we also recognize that FEMA is already working on basically all of our recommendations we would say the the strategic plan that that was from FEMA that covered 2018 to 2022 was a we believe was a big step in the right direction the supply chain resilience guide we also felt that it was basically kind of saying what we have we were saying and really supporting what we were saying in our report and there are a bunch of other things that that are happening within the emergency management and supply chain recovery and resilience arena that are really in that are synergistic with the recommendations of the report so the next slide please and we also understand that FEMA doesn't have the capacity or the maybe the role to to do all of these things that we're recommending but we believe that FEMA has the capacity to provide the leadership to actually to address these issues and help the state and local emergency management agencies across the country to to start thinking about these recommendations and issues so some of the things that we were talking about that we have written about on the report includes how FEMA can provide technical assistance for example maybe collecting some of these data data and mapping critical assets linkages and interdependencies and maybe FEMA can again provide the assistance or the the framework for strengthening coordination with private sector and NGO stakeholders so again just to summarize FEMA cannot be responsible for all of these activities on its own but it can provide the leadership in convening coordinating and empowering others next slide please and finally this is our last slide we also understand FEMA's ability to actually pursue some of these recommendations will be shaped by within the larger policy context and there are certain existing policies and laws and regulations that are in place such as the Stafford Act that will put constraints on the way that FEMA can actually act so one one big statement that we're making is that law makers may wish to consider whether such constraints need to be addressed in the future within the broader policy context thank you Osum and thank you Craig and Stephen as well for giving us an overview of the report itself as you've been noting there have been questions throughout and the speakers have been doing a good job answering those questions throughout as part of the study as Stephen mentioned MIT was doing work in parallel we put together a report in May that was in input into the the study itself and Mike I'm going to acknowledge my colleagues Justin Budali who was a postdoc is now at University of Washington as a professor and Michael Wendell who actually is now joined FEMA directly we had we had come up with some of the you know insights that align really well with all those those recommendations about how information is gathered and then how we can you know that that you've created the recommendation three about sharing information but also recommendation two around building system level understanding and that the systems and the dynamics together and the role that models can play in this so regarding the visibility the private sector has certain ways of seeing things within their organizations but achieving that across public and private requires some more deliberate leadership as was indicated and I think that you know again to the question that Camilla asked about the sharing of information you know the as Osum indicated in the presentation there's there's kind of creating trust in the trusted spaces is important and there are voluntary trusted spaces like the FEMA NBOC national business emergency operation center which is active right now in helping private sector and public sector coordinate and respond to the coronavirus and then there are mandatory trusted spaces where like DHS critical infrastructure sets up you know a formal relationship around critical infrastructure to understand what's going on and I think that a study helps to illustrate that maybe that definition of critical infrastructure should be extended a bit further and I'll maybe brief illustration of that in a moment the other thing is that the models as Osum mentioned can be used for analysis addition support but also to just build intuition about how the supply chain systems work and how to think about the problem in in in a context so next slide I want to walk through a simple example of the deep dive we did on fuel regarding information we looked at different different kinds of data that could be relevant toward helping the emergency management use on the right side of that graph on the left around situational awareness that kind of intuition looking at predicting and forecasting and then the ability to create what we call actionable interventions things that can be done in the midst of a crisis and specifically looking at what kind of data would be useful for a fuel supply chain as a what we call a sentinel indicator of issues that may be emerging so that then action can be taken now in the specific case of Irma we were looking back at the hurricanes in 2017 on the right graph you see there one of the ways that that mandatory trusted space that FEMA has for critical infrastructure they have the ability to collect data and work closely with the private sector to understand the supply of fuel and as he is a phil philip craig philip mentioned earlier that the houston area is very vital to the supply of fuel and you see this in september 2017 after the after harvey the texas gulf coast ability to produce fuel was diminished going from a typical utilization of 90 percent down to 58 percent but you also see that the other regions were able to ramp up supply and be able to fill the gap right so that that refining capability even the even despite the big impact of harvey was resilient next slide we also through the through the work with dhs and ea a contract pipeline the fuel in the pipeline and you can see the areas circled there are the months that were in 2017 during those those hurricanes you can see that the aggregate pipeline stocks as well as the terminal stocks were not severely impacted because of that resilient refining capability but then on the right you see we were able to this is we worked directly with some of the fuel retailers to capture demand at some of their key fuel stations in florida and as craig mentioned earlier that demand spike was a huge thing you know the this is this is more of a daily during the you can see in the box there the state of emergency the evacuation and then the landfall of hurricane irma a big spike ahead of time and then a drop during the during the event itself that was a big shock on the system right that was resilient upstream but that demand surge was a big shock which we can now understand in the current situation is happening at grocery stores next slide please then the question became where is the bottleneck and this was um from a report that the florida dot did that showed and there's lots of examples of where the fuel coming out of those terminals into trucks to get to retail stations was the bottleneck you had lines of trucks waiting to get refueled at these terminal racks to pull the fuel out of the terminals and then get them to the retail stores so our question was how do we assess that bottleneck with tools and so next slide i'll show you a little bit about what we developed at these uh at these terminals a fuel distribution model to look at um uh how we can study that with with some advanced tools next slide so this is what we developed is a visualization that shows the big cylinders there are the terminals themselves and show the relative capacity of the terminals and you can see that in the tampa area there's a huge capacity for terminal this is based on the number of fuel racks to get fuel out of the ground into the trucks jacksonville also has a large amount a large throughput capacity there orlando to some extent there's a pipeline that comes in from tampa that feeds orlando um and then there are some other terminals in the southern part of florida miami and and so forth and one on the on the east coast in canaveral relative you know different sizes of terminals and then the dots are the stations and you can see the numbers there how many fuel stations there are in different geographies next slide we then have the tool that has the ability and it may be hard to see in detail here in the slide but it has the ability to do different levels of demand um as well as for each of the terminal groups you'll see on the top up there there's different terminal groups we can click on and adjust the model to reflect specific demand spikes in certain areas because that was a big issue in florida during ermah that demand surge we could now put in scenarios where we could capture different levels of demand um based on what we're seeing at the time and then there's different different capacities but within the terminals that could could affect the ability the throughput of that and this is based on we actually our team spent some time visiting terminals talking with terminal operators to and with also with the distributors to understand the issues um and i'll point to three key um rates that are in there the the gate the the fill and then the empty the gate is the ability for the trucks to check in and there's a check-in process it's a very um hazardous material with with specific regulations and taxes so they have to each driver has to check in they have to have a card to handle the terminal the hazardous materials at that terminal there's a lot of check-in and and protection for the terminals which is appropriate but that check-in process might be slow and we observe that there's ways to improve perhaps the ability for drivers to check in at the terminal um in a normal situation that isn't an issue but when demand spikes that's when you start to see those those bottlenecks emerge in addition when they when they actually get checked in then they go to the fill the truck right and the fuel fueling capability may be limited by some of the racks only have gasoline some of them only have diesel they may not have both some of the trucks may not have the ability to have several hoses may have three hoses to be able to put in instead of five so there are things you can do to adjust the filling rate perhaps by changing some of the regulations there and then the emptying at the at the fuel stations themselves and also the delivery and then that delivery to the fuel stations being affected perhaps by congestion in the on the roads but also access and the ability to access local communities that are controlled by local authorities so we have the ability to change some of those to build insight next slide please that allows us to then with different scenarios understand what is the demand that's being served and you can see you know and based on the assignments of the stations to the terminals where there is likely to be all the demand being met where there are going to be spots where it won't be met this is one of the kind of medium level scenarios that we looked at but it helps you build intuition as you were to change policies and parameters and even the ability to relocate fleets to different terminals depending on the areas that were the demand discharging we can see how real time you could look at the ability to serve demand the next day and so we could build intuition ahead of time but you could also look at on a day to day basis say if we made some tweaks how could we better serve demand tomorrow and then the next slide is the last one that shows a dashboard that shows the in the upper left the underlying demand the population the middle is the fuel being served and on the right we have a lot of metrics around terminals how they're utilized where the bottlenecks might be occurring in the terminal with this kind of uh uh suddenly uh what it's called a sun i forget the name of it now i might the visual guys in our in our cave uh talked about this kind of a specific way of being able to see bottlenecks shifting with those colors i know we can't do this in real time but just an example of the kind of tool that could help build intuition around a bottleneck that emerges because of a demand issue or a supply issue and being able to diagnose that and address it in real time so let's um let's close down the presentation i just wanted to give you a glimpse of what that kind of tool could look like we will reschedule uh an event where we will demonstrate how this tool works uh at a future time but i think it builds on a lot of the recommendations that the committee came up with about how to build intuition and how to to make decisions in real time to out to meet surge capacity which would be very useful at this time in in meeting the needs of grocery stores and pharmacies um with that um we've had some questions throughout that we've been answering um i don't know if the committee members want to weigh in any further about uh some of the questions that have come in um or any final reflections on this there's no new questions that have been arriving so any final reflections you have uh based on and maybe thinking about the current situation how has this report changed your thinking about what's happening uh in the current coronavirus context jared in terms of the third question that was asked about lead times and and pre uh kind of anticipatory um uh filling up of the of the of the response pipeline we didn't talk about it much but there was a fair amount of interplay between these these three hurricanes given or four hurricanes and uh the three hurricanes and the um because of the because of the the timing and they were so they were just on top of each other and it certainly was clear that the that the um the there there is an attempt um that we saw to anticipate um and when we got to florida and and got to portorico and got to the u.s virgin islands uh it was it was the whole system was just uh i'd call it emotionally taxed um to be uh you almost felt like there was this this can't be true that we're gonna have another hurricane in in three days time and i think there was i think that was really manifest in the in the in kind of how these happen back to back and maybe coming back again to the third question uh the end of the question is about should there be more consideration given to more expensive distribution channels and i would say definitely if there is a need um and i would say again it really depends on the situation on the severity of the situation and and and the and the fact that what would this more expensive channels buy you know in terms of time and capacity again i'm sorry coming from a system perspective most of our answers i think would be on it depends and that kind of you know brings out to the point about these decision support tools and some of the models and looking at the data such as the one that jared just showed us that we would want to look at those uh those kind of tools to make these judgment calls or the intuition that those kind of tools will provide us even without having the full information look have building some intuition based on maybe even some expectations of what happened using those kind of tools will can really help to make those trade-offs between should we go with a more expensive distribution channel or not should we start pushing right now or should we wait see if you have any final thoughts here you're on mute i think just uh i just wanted to reinforce the importance of this kind of engagement across a wide variety of of insights and experiences um from academia as well as the practitioners and connecting directly into the experiences of fema and bringing bringing the sciences into the into practice i think we experienced that throughout this this about throughout the study and also want to applaud fema for quick work that they are doing in acting on these on these recommendations to try to get them in place from moving towards training and better understanding at all levels what how supply chains affect what we're doing i think in our current experiences with the coronavirus there's a much broader understanding publicly about supply chains and their interactions and impacts and while what we're experiencing now has some different dynamics given the very different geographic extent of the impacts from viruses and shutdowns and the potential duration there's a lot that we can learn and take from this study into this experience and others going forward yeah thanks debon and i will note that the two organizations that we're also supporting the study cna and our lab at mit continue to engage with fema they are um engaging with the private sector through the mbeoc anybody that wants to be part of the conversation that's a really good place to connect um and then we're we're trying to facilitate more of that communication and analysis to understand supply chains as they're evolving in the current situation so i as i was on it indicated i think fema is already taking a lot of action based on this report um from what i've seen they're trying to use it to the best they can in this particular crisis um and i think that we're you know this study was really important in putting a path forward that um emergency community can really engage in meeting people's needs during crisis so any final thoughts i think we're close to time here so Jared i just saw another question that's posted on chat and not in the q&a i don't know if you wanted to yeah i did see that too i was wanting to close up here but yeah well first of all one of the commenters said that they got the the diagram i was talking about before was a sunburst diagram so thank you for that um and then there is a question about regulations and tools that local administrators use to discourage hoarding or other undesirable behaviors like price gouging and during an event like this um you have a thought on this Oslo I mean the ideal answer should be yes of course but it's not really clear what that's what kind of a mechanism or a tool we can have i mean we do talk about incentives and i would say the other thing that might help that might help is the right level of information sharing so usually people tend to hoard more when there is no clarity or to the future especially if i don't know if this disruption is going to be holding on for one day versus one month makes a very big difference especially let's say in the current situation it not knowing what the grocery supply chain conditions are i think induces everyone to hoard more because we don't know what we'll be getting in the supermarket aisles in the next week in the next two weeks and so on and so forth so i would say that the right level of information sharing and the reason that i'm using the right level is because sometimes too much information sharing could also induce a lot of hoarding like if you know that a big shortage is coming uh that's maybe not really the best thing to broadcast in an unmanaged way but especially information that explains what's going on and maybe the risk and the and the you know the timeline of what we might expect actually help people to plan accordingly and not to hoard yeah and i will say that um one of the key messages that is is trying to be shared at this moment is here that there is a close coordination with the the private sector with the grocery sector that i've seen um with the government to make sure they're standing ready to support the private sector as needed um but that you know so far the the private sector supply chains have been able to respond and will be able to so that you know that message getting that message out clearly to then shape the demand and those as you saw those demand spikes can be a lot of the reasons that thing that you see shelves empty it's not a not that there's a failure or a lack of capacity upstream but it's just they can't keep up with the demand surge and so if we as a as a community and together understand how to balance the supply chain our consumer behavior as well being a big part of that um then we'll be able to make sure that food gets to the to everyone effectively in a crisis so um anyway well with that i i do hope that uh people continue to stay engaged um in the crisis be looking for the right messaging and sharing that about how supply chains are working um again a voluntary place where where businesses can join is the nboc with fema at this time i know that's a really good place to connect and make sure that you understand how your supply chain can be enabled uh and going forward one of the big things will be um if we have further constraints on access and mobility to make sure that trucks can access communities effectively and any business that it has questions about that as it goes forward nboc will have a lot of information to make sure you know how your trucks can reach the communities effectively even as we may constrain movement due to social distancing so please do stay attuned to those channels that are already set up so that we can approach this as a whole community thank you panelists for taking time uh from your virtual coronavirus office and sharing this uh these insights i think they are relevant for this time um we will post this video as well as the presentation for those to disseminate further but uh thanks committee for all the work on this to help us be better prepared for a crisis like we face now okay thanks bye