PrepTalks: Dr. Jarrod Goentzel “Aligning Public and Private Supply Chains for Disaster Response”


[PrepTalks Theme Playing] it’s a privilege today to be able to
share with the emergency management community, the professionals at the front line of meeting needs and disasters, some of the
things we’ve been learning at MIT. The students and researchers that are in the
group in our center. Emergency managers work tirelessly
before and especially after disasters to identify gaps in communities and
identify responses to meet those gaps at all levels of the government. This
important role relies a lot on supply chains to deliver the resources to
provide the response for those communities. Today I’m going to make the
case that these responses should also incorporate and align with the private
sector, which has far more capacity to act in these communities and do good. And I’m going to illustrate how analysis can help us better align those supply chains
in developing tools and skills in the emergency management community to inform them. I’m going to start by making the point that’s been made by a steady
conducted in the mid-atlantic a few years ago that private sector capacity
is vastly greater than public sector capacity in the disaster. Well it’s
actually both in case of blue sky, the normal operations, and grey sky, after
disaster. There’s a documented case that the supply of food into the
Washington DC metro area over a year’s period 3 million tons is
equivalent to the supply of food over two and a half years into the theater of
Afghanistan and Iraq. So significantly more capacity dealing supplying one city
then is supplying entire supportive operations in those fields. In gray sky
in the week following Hurricane Sandy one distributor was able to distribute
24 million cases, which is 50% more than they normally distributed into that
region. They were able to ramp up significantly to meet the needs, but more
importantly that’s roughly 18 times the volume that the entire public sector and
NGOs were able to provide in a two-week period. And in case you’re not doing the
math, two weeks versus one week, that’s actually 36 times the capacity that one distributor has over all of the public sector and NGO
community. So how do we tap this capacity, and align with it, and make sure that the
response efforts that are identified are not going to get in the way? Well this
starts with preparedness and here’s a picture of the warehouse in Florida the
Division of Emergency Management. And there’s a lot of water in this warehouse.
You see racks with pallets and water, after water, in fact there are 300
truckloads of water sitting in this warehouse that can be deployed at a
moment’s notice to meet needs. You may as well is that an effective use of
government resources to buy water that actually expires in the bottles. You have
to rotate this. If there aren’t any hurricanes or big disasters over a
number of years, you’ll have to dispose of it or donate it. But by
working with the private sector in preparedness work they were able to have
a warehouse full of water that is not actually owned by the government, it’s
owned by the private sector. A water manufacturer keeps its stock in their
warehouse, uses it to satisfy its own customers on an ongoing basis, thus
rotating the stock but also making sure that they will on a moment’s notice
deliver for Florida to meet needs in a disaster. That coordination ahead of time
with the public sector increases Florida’s capacity to respond. FEMA does
this in a slightly different way they have warehouses where they own
their own stock, but they also have contract inventory at warehouses in the
private sector that can be deployed. We did a study a few years ago a master
student to look at the capacity. We start by saying, “Well what is the expected
demand we would need to meet and how can we meet it?” We looked at historical
disasters from a database, tracking and you see the map here that has dots
around the U.S., which locate the center of the disaster or where you would
locate most of your points of distribution or “PODS”. And the size of the
circle is the magnitude or the affected community the number of affected
people in each disaster. And if we were to set up stock to anticipate responding
to that broad portfolio of risks, how would we best do it and what is the
capacity of the stock we have to meet needs? We looked first with though the warehouses that FEMA owns and stock in across the country. And then did a simple measure. This graph shows on the horizontal axis the time to respond up to 72 hours at different points. And this
is the ability of taking that stock, transporting it to those of that broad
variety of potential disasters. On the vertical axis it shows, on average across
all of those disasters how much of the demand were we able to serve? For some of them you’re serving 100 percent, a small disaster you have enough stock to meet
all the need. For some of them you’re serving much less, but on average
if you look at the point at 24 hours, the stock that they actually have serves
about 31 percent of the effective demand on average for that portfolio of
disaster. And it turns out if you were to optimize and move the inventory slightly
to better locations you can improve this a little bit, by 2 percent, up to being 33
percent of the of the affected population served within 24 hours. So we
can improve on this. This is FEMA’s own stock in their own warehouse. Let’s look
at the contract stock, which is available to purchase and to deploy immediately
but sitting in a warehouse in the private sector. There’s three vendors
that were, at the time looking at these contracts for water, and across those
three vendors they added an additional 5 percent at 24 hours, 5 percent more demand can be served with that contingent stock. But what’s interesting is, if you were to
redeploy this the contracts among those three vendors in a better way you could increase that to 17 percent, or by better aligning your preparedness with the private sector you have a much bigger
impact than what you can do in your own warehouses to better meet needs
following a disaster. So we could now move from 36 percent to 50 percent on average across that disaster portfolio primarily by working better with the private sector
in preparing our supply chains. This of course requires some analysis
this isn’t a number we just come up with, and of course I’m from MIT I have to
have my requisite slide with a bunch of math, right? But we take these risk
portfolios of disasters, and by the way we can take an example of hurricane
tracks, we should be using forecasts of the future not just history, any risk
portfolio. We look at the inventory and the suppliers that are available. We also
are now incorporating carriers, not just what inventory is sitting where, but
what kind of contracts and access do you have with the transportation sector
to move things quickly. Looking at all of that, and put it into a big stochastic
linear program, and we put out metrics. Now there’s there’s two things I want to
illustrate here is one we can actually help cultivate some of the skills that
enable the emergency management community to do this. Not everyone in the
community has to do this math, but it’s not that complicated we actually teach
this linear programming approach in the first course of an online micro masters
that we’ve created at MIT. In fact, the first micro masters offer by MIT is in
supply chain management. So there is the ability to build some skill sets, not
everyone, but people can enhance their skills. At the same time we can develop
models that lead to metrics, because in the end not everyone’s going to be able
to run a model or have time to do that and maybe you need to boil it down to
something simple. So I’ll show an example of a metric that we developed a few years
ago internationally and how that’s played a role. It’s something called
the balanced metric and there looking at the sam,e issue the UN depots around
the world the UN Humanitarian Response Depots, house stock for various NGOs and
nonprofits in order to deploy after a disaster, the preparedness stock. They
also every day put a real-time stock report on their website, so we started
downloading this and running it through our model and saying how balanced is the
supply. If they moved it to different places could they improve the response. And we’ve tracked this over time, over the last three years, and on the vertical
axis the number as it goes higher that’s not good that means a number of
one-point-six which is what it was to begin with with one of the commodities
means that you could improve the response by 60 percent one point zero means you’re as good as you can do, right? So there’s some out of
balance, right? They’re not in balance, but over time we’ve been seeing that this
allocation of stock across the community again they’re not directly coordinating,
but we have a metric to see how well we are prepared as a community, that number
has come down for a lot of the commodities. So with a simple numbered we see how well aligned are we to respond to the disasters that are out
there. So turning analysis into metrics is something that we in academia would
like to do with the emergency management community to make it useful for
decision-makers. Now let’s turn to, beyond preparedness, what happens when you have a hurricane approaching? This is the map of Hurricane Irma in 2017 as it
approached. This is Thursday morning, it made landfall on the weekend, so by
Thursday morning most of the private sector companies were already enacting
their plans to continue their operations. For a lot of them that meant taking
trucks full of supplies and positioning them in Southwest Georgia out of the way
of the track, which was projected to go up the East Coast, and ready to roll as soon as the storm passed That’s a good idea but one issue
is, it’s a long ways down into Florida right from Valdosta. To go to Miami one
way it’s 450 miles, to Key West it’s 600 miles. Most trucks can’t make the round
trip without refueling. So those trucks position with supplies often sat waiting
until they had assurance that the truck stops along the way, like this
Pilot Flying J, that this tanker truck is able to deliver fuel to make sure I can
refuel the truck right next to it, which has all the goods that are flowing into
Florida. So to make sure that the private sector can provide for various
important goods, food, medicine, water you have to make sure that the fuel trucks,
the tanker trucks, can get to the stations. and give that assurance to the
private sector that that’s possible. This was an issue in Hurricane Irma. Port Canaveral, one of the places where these tanker trucks fill up, noted that
750 tankers loaded over a 48-hour period. And you can see in that picture on the
left side of the picture you see these fuel racks where the trucks come up and
access the fuel from the big tanks. And you see on the road the line of trucks
waiting to get in to that fuel rack. And those waiting times could last as
long as four hours to access those fuel racks, which meant that tanker
trucks that normally make two, three, or four round trips a day to refuel
filling stations, could only make one or two, which severely diminishes the
capacity to provide fuel through these fueling stations. So what can the public
sector do to help with this situation? There’s a bottleneck here. Well one thing
that can happen is you could avoid the racks. In fact in Hurricane Maria,
right after that, you had normally on the left side you see that trucks filling a
racks, but on the right side there’s the capability of pumping directly from
barges they are bringing fuel in to trucks. So that’s one capability is
avoiding the racks and building special capability. The public sector often has
those kinds of unique capabilities. Another thing that can happen is
providing escorts, well they may wait in line to get to the rack at least when
they get filled up they aren’t going to get stuck in traffic, right? They’ll get
out they can make more round trips, again continue providing fuel to those
stations. Or you could actually start dispensing fuel through your own
agreements. FEMA does a lot of work with DLA, Defense Logistics
Agency, to provide fuel to emergency responders and in fact they
did work with the state of Florida which requested the ability to pump fuel for
emergency responders and for the power restoration crews at community colleges. So being able to set up a parallel network so that those emergency
responders don’t go to the filling stations and take some of that scarce
supply that may be in the private sector. But there’s a question with all of these
things which is the right approach? In fact we wrote a case study, a teaching
case, we had colleagues at Pilot Flying J worked with us to talk about what
they’re doing in their emergency operations center, what FEMA is doing, how decisions are made and how can we better align those decisions with better communication? What could we have done better to align the the fuel supply
chains that enable all the other supply chains to work better? So you’ve got
these parallel networks, we need to align them to increase the overall capacity. And we also need to understand the nature of
that fuel supply chain ahead of time, right? I don’t know how many of you are
aware of how fuel gets to filling stations but it’s refined mostly in the
Gulf Coast and it runs up this pipeline, huge pipeline, up the East Coast. And
those terminals or those fuel racks are all along that pipeline distributing
fuel throughout the East Coast, but notice Florida not on that
pipeline. They receive their fuel through barges that go to ports like Port
Canaveral, which we mentioned before. There’s a few pipelines to the airports
in Orlando and Miami, but mostly it’s that barge, to that truck, and the fuel rack and inland. Understanding how that works ahead of time would enable the emergency
management community to better support the private sector in restoring that
fuel supply, so that all the other good supply chains can be performing. But you
may want to also understanding that whole complex system of fuel may be a too much
to take on. So how do we prioritize the analysis we want to do, the data we
need to collect? How do we determine the key nodes in this fuel supply chain
network, in order to establish relationships with the private sector?
Trusted relationships where information can be shared ahead of time, but
especially after disaster on a regular basis to be able to monitor what’s going
on in those supply chains. Well in this case we can we can learn from
the international community a bit. I’m going to take you to Darfur in Sudan.
This is a project we did in 2012 with the World Food Program. They were doing
exactly that, understanding the private sector capacity in Darfur to deliver
food. The private sector in this case are the people sitting at those chairs, these
are the local traders that buy through other traders and bring in supply to
meet the beneficiaries who are on the other side of the table providing
vouchers given by the WFP. The international community has over the
years and tried to leverage the private sector more and more in providing that
local response, which is more appropriate, and effective, and grows the economy but
they have to learn how the private sector works. They develop some tools one
of these is EMMA, there’s several others that are out there, but this one’s called
the Emergency Market Mapping and Analysis and that’s
exactly what I’m talking about here mapping and understanding the private
sector, and analyzing what it can do, and designing our responses accordingly. And
you see this the three strands in that graphic gaps and responses. That’s what
emergency managers are good at identifying gaps, determining responses,
but there’s a third strand that’s very important. That’s the markets or the
private sector. Understanding their role in this is key and making a more
effective overall response. And in doing this they looked at markets but they
suddenly realize that this market is not just the local market, there’s a whole
lot of other markets that that supply that, and in fact they’ve seen it’s a
supply chain, right? This supply chain very complex, but understanding it ahead
of time is important so they could identify what’s going on. This is the
blue sky, what happens in grey sky? And they’ve got these approaches to try to
approximate this right, but it’s very difficult to gather data in a crisis. And
again we have to have those relationships. Who are we going to call?
If we’re trying to figure out how fuel is getting into Florida, do we call the
fuel racks? That seems pretty obvious. Do we call the barge providers who are
bringing it in from the Gulf Coast? Do we call a few key fuel stations to see how
they’re doing? Who do we talk with, how do we gather data, how do we quickly assess
how this capacity is responding in this disaster? That’s a prioritization effort
where we’re going to borrow from the public health community. We’re developing
some methods now we’re coming up with some approaches in a study sponsored by
FEMA conducted with the National Academies of Science and Engineering to
look at sentinel surveillance. This is used in public health to monitor the
spread of, in this case in Ghana HIV/AIDS. Rather than having monitoring at every
hospital and clinic they pick a few key ones to see what’s going on overall,
because you can’t gather that much data and analyze it, right? So they have methods for picking health facilities to monitor the health of a community. And they realized that they also have to understand relationships of
people in spreading diseases, so networks come into play. These are all very useful
because we want to look at how we monitor the health of a supply chain, and
these are networks. And we to pick these sentinel points in a
network. What are the sentinel locations we need to establish relationships and
have communication with the private sector, in order to understand how
they’re coping with the crisis, what capacities they have, and where they need
help. This prioritization will enable us to
not only help fuel supply chains but all the various supply chains with their
tiers of suppliers and tiers of distributors. They’re very complex every
commodity has a complex supply chain that we need to identify sentinel
indicators in, to establish relationships between emergency management community and the private sector, to together align the efforts in responding to needs. So in
conclusion I hope you are convinced as much as I am that the private sector has
a lot more capacity to respond, and that there’s a new role that emergency
managers can adopt in understanding that private sector capacity, and helping
to determine how to help restore those operations that enable the local markets
to be filled up, the local stores to be filled up, and the economy to recover
quickly. And that we need analysis to do this. And that analysis is a combination
of developing new skill sets, but also developing new tools and approaches, and
we are in academia ready to work with you to develop the tools that will make
your job easier in aligning your capacity with the private sector and
meeting more needs after a disaster. Thank you. [Applause] It’s a good question prioritization is
always key. Well you can tell by my talk fuel is a pretty
important one. I mean and it’s back to the interdependencies that was mentioned in earlier talk today, that fuel enables power, right? Power with your generators.
Power enables fuel to get the pumps out of the ground, so there’s a lot
of things, but power and fuel I mean supply chains need that. We got to have fuel for the trucks to roll, so I think fuel is an obvious one, right? And
then you look at the priorities that that are really that a community will
need in that short time window, because you should really focus on the supply
chains that are most urgent, you know so you’re looking at your food, and your
water, and your medical, right? And then and diving in, and sometimes diving in more specifically with certain commodities you know might be an issue. And this may be contextual in a certain region where you know things are going to be more critical than other. So I think it comes
to aligning with what the priorities are in general in responding, and then
identifying those private sector supply chains that provide those critical
commodities, and then starting there. But we’re going to hopefully come up with
some better tools over the next year in our research to help make that process
easier. It’s an interesting mix because some of
this is you need to understand a national capacity, and a lot of the big
providers of food and medical supplies and so forth are national,
right? So you understand what their capabilities are nationally and they’ll often design their supply chains so that the national can be a support
for the local. But the most important information is how the local
supply chain is actually working. So you need to have both levels, and in fact I
would argue that the the the best information comes from that local source.
Having the local emergency managers to be able to have that discussion with the
operations locally, then they can run that up nationally, but also then the
companies can do the same thing with their local store manager and so forth
running it up. But they’re also is a complementary effort, you know not
every community can be gathering this kind of data, and they’ll need support
and, tools and so forth. And so I think there’s a combination. In fact
internationally the cluster system is used to to help coordinate and
monitor disasters. And they have a set up where they have the local clusters in a
disaster, but they have a global cluster that helps enable capabilities on an
ongoing basis, to do training on how to engage with the private sector and
understand its capacities and leverage it, but also to help gather data about
capacity that may be difficult for any one jurisdiction to do. And I’ll say one
other thing too companies don’t look at a community in a
disaster the same way as the government does, because the government has its
inherent jurisdictions of counties, and states, and regions that may not define
what a market would be for a company. And so there’s also a misalignment in terms
of jurisdiction and framing the nature of what that community is. So
there’s some challenges in aligning our way of thinking, but I think
you have to have the local level engaged because they’re going to have
the best information, with support at the national level on how to
train and do that. I’m glad someone asked about the micro
masters program. It’s been one of the more surprising and exciting things
that’s happened in our Center and I’d say in our University. I mean seeing that there was this desire for learning out there, and that access to it was kind of a challenge, and being able to scale it up
and be able to teach thousands of people at a time. We’ve had hundreds of
thousands people take our supply chain courses, we wouldn’t have guessed that.
And I don’t have a number on how many emergency managers are doing it, I hope
more do it after this talk. But we definitely are trying to make it so
that we’re teaching the supply chain and analytical tools, and the design
tools, and the strategies that we teach our students in our graduate programs
who become professionals and who are really good at running supply chains. And
so it’s a fairly advanced curriculum and you become a real expert in how to do
this. Not every emergency manager needs to do that. And in fact so one thing I’m
doing, I’ve been teaching a course on humanitarian logistics for a number of
years on campus. I’m now working to it’s a lot of work to create these online
courses, but I’m putting the effort out because I want to make this kind of
information available. And my course is going to be designed more for a broader
emergency manager to understand the nature of how a supply chain is managed,
not necessarily do all the analysis because if they’re really interested then go on and continue doing the master’s program and
develop those skills, but at least be able to map to the approaches that are
that a supply chain professional uses to manage a supply chain. But with the
context that changes a lot of things from what the private sector normally
faces, and how does that context change the way we formulate our problems, and do
our analysis? I’m working to develop a course to offer that, so I I hope that
that can play a role in in helping to improve the ability for the emergency
managers to understand and analyze the private sector. Yeah, this an important question. I think Maria helped illustrate this a lot because there’s a lot of critical
manufacturing on the island of Puerto Rico. And that well we’re focused on
bringing things to the island to help the people that are affected, we have to
make sure that things can come back out island and that they can they can
continue running operations. I would say we convened calls with companies in the private sector who were trying to make sure that they had their
operations you know up and running as fast as possible. Power was a big issue there. Looking at that, being able to have the facility up, but as was mentioned by Yossi in the previous talk, people. Making sure
their people are taken care of and so you have to have a workforce to run your
operations, you have to have power to run your operations and then one of those
once those things happen then you can can restore that. But understanding the role of manufacturing in a broader and how it serves
broader needs in the U.S. will help us also prioritize which facilities may
be more important. Every facility is important, we want to get the private
sector running as fast as possible, but some may be more critical than others. And especially if we understand more about those supply chains. You know
they may not the obvious ones there might be a second tier supplier
that produces something you don’t obviously see. A good example is fuel
filters, was something that I discovered was an important commodity. I would prioritize this, why? If you don’t have a fuel filter on a generator the maintenance is much more significant, right? So you if you
don’t have fuel filters, then you’ll have a lot more maintenance issues, generators
going down, you won’t have power. So little commodities like that, you know
identifying which ones in these complex systems, as was mentioned earlier today,
being able to assess which commodities are most critical and then identifying
that manufacturing footprint and the transportation footprint, there may be
unique transformation resources required as well to move the goods, will help that
prioritization of that resource allocation. So it’s important to think about the response and recovery and there’s not a clear line of how you transition, and
what that transition is and it will depend on the context. You know the
Hurricanes in 2017 you had different times of response and recovery for each
of those, right? And so I think part of the thing is to identify when are we still responding and is it more of a crisis, and when are some things restoring to some extent, and now we can think about how to
recover and build back better for the future, right? And part of this is
monitoring some of these key supply chains and understanding how well
they’re performing, how well our needs being met in the community. And there’s
been a lot of indicators developed of community health and looking at that
whole community, but the supply chain and I’m specifically looking at how we help
restore the supply chain and make it work better, but also help it
recover to be more resilient the future. And I think by having a better
understanding of how the supply chain works in the blue sky, and then being
able to pinpoint the most critical points in gray sky will help us
determine both the ways to enable the response, but also how to build it back
better harden the parts of the supply chain that need that in order to recover
and be ready for the future. But I’ll also add recovery is a lot
about economic recovery and the sooner you get the private sector back in that
economic recovery starts happening. People can go back to work, kids can go
back to, a lot of good things happen if you can get that private sector and
those markets working again quickly. [PrepTalks Theme Playing]

You may also like...

Leave a Reply

Your email address will not be published. Required fields are marked *