July 16, 2024 • 47 mins
Vehicle routing is one of the most well-understood, extensively studied problems in both history and academia—it's been studied by academics since the early 1800s. Yet 200 years later, optimal efficiency remains just out of reach. And given the enormous increase in pressure for consumer expectations in recent years, that remaining "last mile" of vehicle routing efficiency could, on a global scale, make a huge difference to a huge number of people.
In this episode, we're joined by Matthias Winkenbach, Director of Research at the MIT Center for Transportation & Logistics. Matthias is an expert in urban logistics, last-mile delivery, and vehicle routing, and he has just launched a new lab, the MIT Intelligent Logistics Systems Lab, that will use AI and machine-learning techniques to tackle today's vehicle routing challenges—and that could make a major impact on vehicle routing solutions where traditional methods and algorithms have come up just short.
Host & Executive Producer: Benjy Kantor Marketing Writer & Producer: Dan McCool Sound Editor: David Benjamin Sound Audio Engineer: Kurt Schneider
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
(00:00):
- Before we get startedwith today's episode,
we have some exciting news to share.
We're pleased to welcome a new member
to the MIT Global SCALE Network,
the United Kingdom SupplyChain Excellence Center.
This newest scale center has just launched
through a partnership between
the MIT Center forTransportation and Logistics
and Loughborough University in the UK.
The UK Supply Chain Excellence Center,
based in Loughborough, willoffer a master's degree program
(00:22):
in supply chain management
based on the curriculum developed at MIT
and will soon begin admitting students
to its inaugural cohort,the class of 2026.
To learn more, visit scale.mit.edu.
(lively music)
Welcome to MIT "Supply Chain Frontiers",
presented by the Center forTransportation and Logistics.
I'm your host, Benjy Kantor.
(00:44):
Each episode of "Supply Chain Frontiers"
features center researchers and staff
or experts from industry
for in-depth conversationsabout supply chain management,
logistics, education, and beyond.
Today, our guest is Dr.Matthias Winkenbach,
a principal research scientist
and newly appointed director of research
at the MIT Center forTransportation and Logistics,
as well as the director ofthe Megacity Logistics Lab,
the Computational Analytics Visualization
(01:05):
and Education Cave Lab.
And soon, thanks tofoundational support from
and collaboration withMecalux International,
the newly launched IntelligentLogistics Systems Lab,
all of which are underthe purview of MIT CTL.
First, this episode of MIT"Supply Chain Frontiers"
is brought to you by
the MITx MicroMasters® Programin Supply Chain Management.
Gain an end-to-end understanding
(01:25):
of supply chain managementover five courses
and a final comprehensive examrepresenting the equivalent
of one semester of coursework at MIT.
Boost your skills at workor pick up some knowledge
as part of your lifelong learning.
To learn more, visitmicromasters.mit.edu/scm.
Today, vehicle routing isone of the most studied,
if not the most studiedissue in logistics,
(01:46):
stretching back about 100 years.
But despite such attention
and major strides in the decades since,
optimal efficiency in vehicle routing
and last-mile logisticsremains just out of reach.
But the comparativelysmall amount of ground
that's left to coverstands on a global scale
to have an enormousimpact on supply chains,
and AI and machinelearning might be the keys
to finding the remainingpieces of this puzzle.
(02:07):
It's not as simple as gettingfrom point A to point B,
and Mathias Winkenbach is here
as an expert in urban logistics,
network design, data analytics, and more.
And he's here today with us
to discuss his very exciting work
with using AI and machine learning
to fully optimize vehiclerouting and last-mile delivery.
Matthias, welcome to"Supply Chain Frontiers".
We're so happy to have you.
- Thanks, Benjy, for having me.
- Let's start with some context.
(02:27):
Vehicle routing is one of the most
studied problems in logistics,
but also one whose solutionis the most elusive.
Can you give us a briefoverview of, as best you can,
the last 100 years orso of vehicle routing,
or if you even can go that 100 years?
- Sure. I mean, the thing about
the vehicle routing problem,
and that's actually why wealso focus our research on this
with these new methods on the horizon,
(02:47):
is that it is one of themost well-understood,
well-studied problems inindustry but also in academia.
So if you look back,the foundational basis
of vehicle routing problems
is called the traveling salesman problem.
That's a problem that academics
have basically been studyingsince the early 1800s.
The vehicle routing problem,
which we today mostly know and study
(03:10):
in the context of logisticsand transportation,
that was first formallyintroduced in 1959, I believe,
by a researcher called Dantzig and Ramser.
So it's been a while since we'vebeen studying this problem,
and so the state of the art in this space
is also pretty advanced by now.
So there are a lot of reallygood algorithms, models,
(03:31):
to capture a lot of thereal-world complexity
that comes along with real-worldvehicle routing problems,
but the way they are solved today
still struggles with basically
getting the last, let's say, 20% right.
And those last 20% didn'treally matter that much
for a long time, when we were in a world
(03:51):
where people didn't shop online,
and transportation did have enough time
to basically get stuff from A to B,
get stuff from a warehouse to a customer
or a shop or a factory.
But in the last couple of years,
the industry has seen somuch increase in pressure
from consumer expectations.
Things had to move faster, hadto be more customer-centric,
(04:13):
had to be more individualized.
And suddenly, for these highlyresponsive logistics services
that people are asking for today,
those last 20% matter a lot.
And that's where existing methods
tend to reach the boundaryof what they're capable of.
- Now, I first heard the term last mile
with regard to people moving, commuting.
(04:35):
I worked for and I was a member
of a public bike-share system,
and it was around 2014 or 15 or so.
Is that the kind of thingyou're looking at also,
or is it specificallycommercial, industrial,
the things like Amazon gettingyour package to your door
as opposed to getting you to work on time?
(04:57):
- I mean, there are certain similarities,
but what makes last-mile logistics,
so basically the commercialmovement of goods
on that last mile, particularly complex
is that we are dealing withwhat we call fragmented routes.
So if you are a commuter,
you're usually an individual
who has to get from point A to point B.
(05:17):
And yes, you're making routingdecisions along the way,
but in a way you're just trying to find
the shortest or mostconvenient or fastest path
from point A to point B.
While someone likeAmazon, UPS, you name it,
they have to deliver to multiple customers
along one consolidatedand efficient route,
so think of the big brownvan on the road out there.
(05:40):
That van typically makesaround 120 stops a day,
and those have to be sequenced in a way
that maybe you're not alwaysable to take the shortest path
between any two customersalong that route,
but your goal is tooptimize the overall route.
So to, for instance,find the overall fastest,
cheapest, shortest paththat connects all of these,
(06:03):
in this case, 120 pointson the map, basically.
- And when we're talking about
the full efficiency of that last mile,
and when I say efficiency Imean sort of all these things,
like the cost and the speed
and whatever the other factors are.
Why has it been so elusive to get to that,
the thing that makes it work completely?
(06:25):
- There's a few thingsthat make this really hard.
One is it's not just a mathproblem; it's a human problem
because one major source of uncertainty
in routing is, for instance, the driver.
Whether the driver isvery experienced or not,
whether the driver is having a good day
(06:46):
or a bad day that morning
is also a factor of how other participants
in a mobility system likea city, for instance,
interact with that one route.
So, for example, companies like UPS
can optimize the heck out of a route,
but then what happens if unexpectedly
(07:07):
there's a traffic accident?
And that completelychanges your assumptions
around travel speeds,
travel times from onecustomer to the other.
So there's a lot of uncertainty,
a lot of factors thatare very hard to control,
if not impossible tocontrol for the company
when it tries to findthe optimal solution.
And that's where, let'ssay, traditional solution
(07:28):
approaches to thisproblem tend to struggle,
because they're eithernot capable of capturing
all those uncertain influences,
or they're not kind of efficient enough
to quickly respond to changesin the operational environment
to basically find the next best solution.
- And I know you've mentioned in the past
that the complexities around this
(07:50):
are based a lot in density.
Now, if we're talking about density
we're talking about population,
are we talking about infrastructure?
What are those things thatare standing in the way
that are challenging,
or that create theproblems in the first place
that aren't necessarily standingin the way but they exist,
and so you have to figure it out?
- So, I mean, there's a reason why
that one lab that I've been leading
(08:11):
for the last 10 years roughly
is called the Megacity Logistics Lab,
Because when we startedout doing this research
we very much focusedon really large cities.
Why? Because that's wherea lot of people live
But, also, that's wherelogistics processes
are the most complex.
And one of the main drivers
of that complexity is indeed density,
So every city, or let'ssay most cities out there
(08:34):
weren't planned for the size
and the amount of peopleand the amount of goods
that need to move into andout of these cities today.
Most cities out there grewhistorically, and that's okay
but the problem is that usually
transportation infrastructuredoesn't grow proportionally,
because it was never designed
to really be able to dealwith that much traffic,
(08:57):
that much demand for this limited space
on the road, basically.
And that's where, as cities grow,
as we see ongoing urbanizationall around the world,
density in these cities increases
and that means traffic density increases,
demand density increases.
And that, again, has bothpositive and also negative effects
(09:18):
on logistics processes.
The positive thing is,
if your customers aremore densely co-located,
then at least theoreticallyyou have to travel less
to meet the same number of people
and to deliver the sameamount of goods, for instance.
So technically, your delivery routes
should get shorter, faster, cheaper.
But the downside of itis, as I said before,
transportation infrastructuredidn't grow proportionally
(09:40):
so you have more traffic,more uncertainty,
generally mobility decreases,travel speeds decrease,
so that makes these routesnot just less efficient
but also less predictable
because there's a higher likelihood
of a random incidentlike a traffic accident
in a high-density environment
as compared to, let's say,a low-density environment
like rural Massachusetts.
(10:03):
- I'm glad you brought upthe Megacity Logistics Lab.
I'm kind of curious, can you tell us
how that lab fits into whatthe current methods are,
what the new methods are,considering it's only 10 years,
it's only been around for 10% of the time
that this problem has been focused on.
And so what are the new methods
for looking at these older problems,
(10:24):
and how do they differ from
what people were trying to do before?
- Well, I mean, when I startedresearching this space,
which was shortly beforethe Megacity Lab started,
but that was basically at the onset
of widespread e-commerce.
That's when Amazon juststarted operating, basically.
That's when people kind of did realize
(10:44):
how important the last mileof any logistics process
might become in the future,
but probably nobody reallyhad it on their radar
that today we're talking about services
like order now and get itwithin the next two hours
or something like that.
And so the traditional methods
that also my labpredominantly has been using
(11:07):
over the last 10 years
can be generally classified
as what we call operationsresearch methods.
So this, broadly speaking,means optimization methods.
And here we mostly distinguishbetween exact methods,
so these are typically mathematical models
where you formulate, let'ssay, a routing problem exactly.
(11:27):
So you write it down as a set of equations
and then you're using algorithms
to solve those equations exactly.
So, ideally, at the end of the day
you want to get the optimal solution.
The problem about these methods is
that as soon as you applythem to real-world problems
that are of a real-world size,
like for instance those 120stops of the big brown van
(11:48):
that I mentioned earlier,
those methods take a very long time
to find that optimal solution,if they can find it at all.
So they're not efficient enough
to really be applied at scale, typically.
That's why the other family of methods
that fall under kind oftraditional operations research
would be heuristics andmeta heuristics approaches,
(12:08):
which is basically a set of rules,
a set of kind of decision procedures
that you put into an algorithmthat are often tailored
towards the specifics of the problem.
So for instance, in thecontext of vehicle routing,
you can encode certain rules
that define what is agood or a not so good move
(12:32):
of a vehicle from one customer to the next
based on your knowledge of the problem,
based on your domain expertise.
So these methods tend to be faster,
tend to be more efficient infinding really good solutions.
The downside is that theytypically can never guarantee
that they find the optimal solution.
(12:53):
So that's just kind ofa two-minute summary
of what those traditional methods are.
And my lab has been veryactive in publishing papers,
working with industry on using
both of these types of methods extensively
on solving all sorts oflast-mile logistics problems,
not just routing.
So the reason why we are nowexploring a new set of methods
(13:19):
inspired by all this generative AI
and machine learning stuffthat's going on out there today
is that, as I said before,the world is moving faster
than these traditionalmethods can keep up.
So for instance, being able tooptimize the logistics system
for services likeon-demand one-hour delivery
(13:39):
means you are optimizing a route
with very little time at hand.
And then some of these existing methods,
some of these traditional methods
might just not be fast or efficient enough
to really solve your problemwithin that little time.
Similarly, the trend towardsmore customer centricity,
so really focusing logistics processes
(13:59):
around the individual requests,
the individual needs ofindividual customers,
exposes them to so much uncertainty
that, again, these traditional methods
have a hard time capturing all of that,
capturing the uncertainty andthe real-world complexity of,
let's say, a modern-agestate-of-the-art routing problem.
And that's where we need approaches
that are more data-driven,
(14:21):
that learn more fromreal-world operational data,
that learn more aboutthe real characteristics
of individual customersor individual drivers,
or individual routes.
And these data-drivenmethods can't really be found
in the traditionaloperations research space,
but are very prevalent in whatis currently being discussed
as machine learning and AI methods.
(14:43):
- I'm really glad youbrought up that idea of needs
and how efficiently someone might need
to have a package delivered to them,
or how quickly a companyneeds to have delivered.
And I'm gonna come up witha more challenging question
for later in the show,
but something else that you touched on is
as you are working more and more with AI,
machine learning, andpredictive analytics,
how is that AI differ from simply someone
(15:06):
or a group of people analyzing data
that they already know exists,
and making a prediction based on that?
- That's the thing, Imean, there's been research
on predictive analytics for decades.
And I'm also not sayingthat this will become
irrelevant in the future,
but to give you a very tangible example,
(15:27):
if you're looking atdemand forecasting today,
there are very capablemethods to forecast,
let's say, the demandfor a certain product in,
let's just say, Amazon's assortment
at the scale of an entire city
for maybe the next day or the next week.
So you can probablyrelatively accurately predict
(15:49):
how frequently that product will be bought
in the city of Boston next week.
But if we're talking about
customer-centric logistics processes
with very short lead time,
let's stick to the example ofon-demand one-hour delivery.
Being able to predict how often
that product will be bought next week
is pretty much useless for you,
what you really need isthe ability to predict
(16:12):
at a very high spatial butalso temporal resolution
when and where that productis going to be bought.
So basically, what youneed to be able to do
is to have a good estimate
of how many items of thatproduct will be bought,
let's say, in a zip code in Cambridge
(16:32):
within the next 30 minutes.
And that's where those traditional methods
that rely on a lot of data tomake kind of point estimates
or distribution estimatesare not that useful anymore,
because you need tools thattake much more data into account
but also detect muchmore interdependencies
between the various data points
(16:54):
that you use to train these models,
to be able to make thesemore fine-grain predictions.
That's one difference between
what used to be just prediction
versus what we now call
machine learning-basedpredictive analytics.
But the other thing thatI would like to point out,
and that brings us back to routing, is,
for instance, up until recently,
(17:15):
predicting a really good vehicle route
was not something that wasconsidered really feasible
because, let's say, if you are to predict
a good solution to a routing problem,
you're not just makingone point prediction,
you're not just predicting onevalue far out in the future,
you're basically predicting a solution
(17:36):
to a sequential decision problem.
So you're predicting asequence of stops of a vehicle
that has to make sense.
Not just one stop that has to make sense,
the whole sequence ofstops has to make sense,
it has to be a solutionthat's both feasible,
that's even viable to execute on,
and at the same time should be
(17:56):
as close to an optimalsolution as possible.
And that's something where,
let's say, traditional prediction methods
were just not capable of doing that.
Now, with the advent of stufflike transformer models,
so actually more or lessthe same kind of methods
that also power things like ChatGPT
and all of these chatbots,
even though they areusing these methods in,
(18:17):
let's say, the context ofnatural language processing,
these types of methodsalso have a huge potential
for optimization problemsin the logistics industry,
like the vehicle routing problem.
- Something funny that cameup recently in a conversation
I was having with Jose Velasquez,
with regard to some Amazondelivery and last-mile stuff
was that some of the thingsthat seem very intuitive,
(18:38):
for instance, that you mightthink that it is most efficient
to deliver packages in a way
that it has the best geographical routing,
but actually that mightnot be the most efficient
in terms of sustainability,and even speed,
because maybe it has to dowith how heavy the packages are
and things like that.
- So another example in that space is,
and something that we alsoexplored a few years ago already
(19:01):
with folks like Amazon and others,
is you wanna come up with route plans
that are not just short or cheap or fast,
but that a driver can actually execute on
and that a driver actuallythinks is a good route.
And here's one of the big benefits
of these more data-drivenlearning-based approaches
(19:22):
as compared to just runningan optimization model,
because a driver, especiallyan experienced one,
knows exactly how to operate a route.
And if you look at manydrivers out there today,
they get a route plan in the morning
and they make their own adjustments
because, for instance,they know much better
than any routing algorithmswhere to park safely,
(19:43):
where to basically avoid traffic
at certain times of the day,
or they might even know which customer
might be more or less available
during which times of the day,
even though that customer may have never
explicitly voiced a preference for,
let's say, being deliveredbetween 10 and 11 AM,
but the driver knows this.
(20:05):
And that kind of consistency,
that kind of tacit knowledge
that's ingrained in every single driver,
is basically impossible to encode
in a hard optimization algorithm
or a set of mathematical equations.
But it can be learned from observing
the actual behavior of drivers over time,
(20:26):
so we may not necessarilycome up with shorter
or faster or cheaper routes,
but we may come up withroutes that are more reliable
because the driver is actually able to
act on the route plan that hegets in the morning, or she,
and also some of hisor her tacit knowledge
about that route is alreadyembedded in the plan.
(20:48):
- And is that gonna changeor has it been changing
the more there is autonomous delivery,
or drone delivery or things like that?
- I mean, today, for better or for worse,
there is not that muchautonomous delivery yet,
especially not in urban contexts
but let's say if we wereever to get to that point,
what you wouldn't wanna lose
if you were fully automatingthe delivery process
(21:10):
is that tacit knowledge about,
for instance, individual customers
or safe and availableparking spaces and the like.
And so you would wanna have
some sort of trained modelrunning in the background
that tells that autonomous vehicle,
or drone or whatever it might be,
how to best deliver theproduct to your customer,
(21:31):
even though there is nohuman involved anymore.
So in a way, if we want these systems
to be as close to a humandelivery experience as possible,
then sure, this is the kind of method
that we should explore further.
- And when we're talkingabout route optimization,
have you seen trends thatcompanies are prioritizing
when they're optimizing their routes?
(21:52):
Have you seen those shift from the past?
Are there things thatyour teams are working on
that you've seen start to be implemented,
and how is that changing the landscape?
- I mean, I think that theunderlying routing problem,
I mean, as I said, that hasbeen studied for decades.
The fundamentals of that problemdon't necessarily change,
(22:14):
but what has definitely been a trend is
how can we focus our solutionsto the routing problem
more around the specificneeds of individual customers?
So, while a few years ago youtreated every single customer
more or less as a uniform mass,
now it's really moreabout individual requests
(22:36):
and individual characteristics.
And that obviously also has implications
for the way companiesapproach these problems.
They have to capture morestop-specific information,
for instance, how muchdoes it actually take me
to deliver to customerA versus customer B?
And that may depend on thenature of that customer,
(22:56):
the nature of the buildingthat the customer is in,
and many other factors.
But, also, we need to understand
preferences more explicitly.
And a few years ago peoplewould opt for something
like time window delivery,
and you would get immediateinput from your customer
as to when they wanna be delivered to.
(23:16):
Nowadays, the expectation is more like,
"Well, the logistics processshould already anticipate
my availability, my preferences."
So even though I don't voicethese preferences explicitly,
your routing algorithm should probably
already take into account
what we have learned aboutthat customer over time,
and therefore basicallymake that customer happier
(23:37):
by, under the hood, alreadycustomizing the delivery
to his or her preferences.
So that shift towardsmore customer centricity
comes in various shapes and forms
but has definitely been
the most important driver of complexity,
but also innovation in the routing space.
- I'll bring in my morechallenging question here,
which is talking aboutcustomer preferences.
(23:59):
If I order a soccer ball online,
my preference would be
that my soccer ball'sin my hand right now.
How do, I guess, we balance the,
there's one part that'sjust sort of like...
Again, the logistics of gettingthat into somebody's hand
as fast as they want it versusas fast as they need it,
(24:22):
I guess, is the question,
which is like theprioritization when it comes to-
You know there's labs at CTL
that deal with sustainabilityand humanitarian,
and if delivering that soccer ball
is gonna impede my delivering water
or baby formula or something like that.
Is that sort of, for lackof a better term, morality
in decision includedin how machines learn,
(24:44):
how people look at it, howcompanies put that into effect?
- Let's say it could be,but it's something that,
let's say, we haven'texplored the morality
of vehicle routing problem explicitly yet
because that's honestlya rather subjective term.
It comes down to the same problem
that many of those autonomous
(25:05):
vehicle manufacturers have today,
like, how do we teach a vehicle morality
if it has a choice between, I don't know,
having an accident that,let's say, kills a child
versus kills an old lady?
What's better, what'sworse? We don't know.
And as humans, if there's a human driver,
we usually don't really have a choice
because it just happens so fast.
(25:27):
A machine might actually have a choice,
so how do we teach it ethical behavior?
And that's something thatwe are probably not equipped
to do in our lab, and that's why we try to
stay away from that.
But what you can do,obviously, as a company is
you can prioritize certain objectives
in your decision-making.
So for instance, if it's maybe
(25:48):
not a life-and-death decision,
but if it's about reducing youroverall emissions footprint
as a whole, as a company, let's say,
if an e-commerce platformwants to make sure
that it still meets thecustomer requirements
as good as it possibly can,
but at the same time tries tobe as sustainable as possible,
then that may be part of that
(26:09):
rather complex set of objectives
that you can "teach"to a routing algorithm
so that that algorithm probably already
makes that distinction ona product level and say,
"Okay, if I have todeprioritize certain products
because either I run out of capacity
or because I'm hitting myemissions threshold, basically,
(26:31):
then which one should I deprioritize,
the soccer ball or the baby formula?"
And so in a way, that'sanother opportunity, I believe,
where these algorithms alsohave a unique strength,
which is learning from customer feedback.
If you get your soccer ball afew hours or maybe a day late,
(26:52):
you might not be thrilled about it
but it's also not the end of the world.
If the mother of a newbornchild runs out of baby formula
because the delivery is getting late,
she will probably be much less thrilled.
And if you can capturethat feedback somehow
and basically use that as your input,
from which you can thenlearn or train an algorithm
(27:14):
to kind of check itspriorities a little bit,
then next time you have that problem
that algorithm is already better equipped
to anticipate the kind of discomfort
or disutility that it is causing
by deprioritizing the wrong thing.
So in that sense, rather than you and I
(27:35):
who design an algorithm havingto make an explicit choice,
whether it's the formulaor the soccer ball,
we can actually havethose algorithms learn
from the response of theentire kind of social system,
which is, in this case, yourcustomers, to your priorities.
And over time, those wouldthen hopefully match.
(27:55):
So what your algorithm does
matches both your corporate objectives,
as well as the general norm
or the general prioritiesof your customer base.
- Yeah, I appreciate that.
I don't know that thequestion has an answer,
just kind of curious when talking about
the different sort of research aspects,
especially within CTL,
just thinking aboutthose kind of priorities
(28:18):
and what kind of goes into this.
Shifting gears a little bit,
I would love to hear youtalk about this new venture,
this Intelligent Logistics Systems Lab,
where the research goes from here.
Is this a step fartherfrom what the other labs
that you're directing are doing,
or is it a completelydifferent type of research?
- I mean, the formation of the
Intelligent Logistics Systems Lab
(28:39):
is basically born out of anobservation that we've had
for several years now,
namely that more and more of the research
that my other labs at CTL, butalso other labs at the center
and other places at MIT, are doing,
more and more of that work
moves away from basicallythose traditional methods
(29:01):
that we discussed before,
those traditionaloperations research methods,
and move further towardsthis "new family of methods"
in the machine learning and AI space,
and most importantly we see itas a huge potential for work
kind of at the intersection of the two,
basically hybrid methodsthat basically try to combine
(29:22):
the best of both worlds, thebest of traditional OR methods
and the best that those newML and AI methods can offer.
Because what we see in ourearly research already today
is that actually the greatestpotential for improvement
over what we've been doingfor the last couple of decades
doesn't necessarily lie in
completely replacing those methods,
(29:43):
but rather kind of putting them
on machine-learning steroids,if you allow that term.
And that's kind of anexciting observation for us
and also kind of goes beyondthe original scope of,
for instance, the Megacity Logistics Lab
and many other labs at the center.
So I think it's the righttime to actually have
(30:04):
a dedicated group at the center
that bundles our activities
in that space a little bit more,
and really envisions kindof the future generation
of logistics systems thatenable logistics services
that we probably todaycan just kind of imagine,
but that will have to become a reality
(30:25):
within the next couple of years,
because that trend towardsfurther customer centricity,
increasing pressure of sustainability,
increasing needs forsupply chain resilience,
all of these trendsrequire us to build systems
that are smarter in the sensethat they are more agile
(30:45):
in responding to changes inthe operational environment,
in the socioeconomic environment,
in the geopolitical environment,
and that are more data-driven.
That are basically, as I said before,
learn from the very rich knowledge
that exists in many individuals out there
in the logistics industry,
be that the driver, bethat the warehouse worker,
(31:05):
be that the supply chainlogistics executive
with 40 years of industry experience.
We don't wanna lose the human expertise
in designing those future systems,
but we need to be ableto scale that expertise
and basically ingrain it
in the mathematical models that we build,
(31:27):
such that we can solvethese very challenging
future logistics systems design problems
at scale and rapidly.
- This is something that I've asked
a few of your colleagues about their labs.
Is there an end goal for the lab?
Is there a point at which youhave done what you need to do,
you've antiquated the lab in a good way
(31:49):
because the solution has been found?
- Yeah, I mean, in a waythe Megacity Logistics Lab
might be a good example for that
because it's been around fora bit more than 10 years now.
We've done a lot of great work
but the lab has to someextent outgrown itself,
so we no longer just focuson last mile logistics
(32:10):
in the very narrow sense,
we are basically alreadyworking on a lot of stuff
that involves more thanjust the last mile,
because today the way youoperate a last mile vehicle route
isn't a problem that is isolated
from everything else in the supply chain,
it's basically highly interconnected
with aspects of networkdesign, inventory, planning,
(32:33):
sourcing decisions, long-haultransportation decisions.
So supply chains and logistic systems are,
or have become much more interconnected
so, basically, focusingon just that one aspect
of last mile logistics may nolonger make that much sense,
at least in many cases.
Which is one of the reasons why,
(32:54):
I mean, the lab will still exist
but I think our focus will probably shift
towards looking at supplychains and logistics systems
at a slightly higher level,
which is in part enabledby also looking at
a broader set of methods
to basically not just optimizethat single vehicle route,
(33:14):
but ideally to find an optimalsolution to an entire system
that may consist of manydifferent vehicle routes,
1,000s of customers, large networks,
many interconnected decisions.
And ideally we want to find ways
to solve these problemsin an integrated way,
so just focusing on last mile
(33:35):
in this case might notnecessarily be what we need to do
to design the next generationof logistics systems.
- We talked a little bit earlier
about sort of the satisfactionfor an end consumer.
What are the implicationsfor everyday consumers?
Again, we talked a little bit
like delivery of the soccer ball, but...
- I mean, the obvious benefit
to the, let's say, the end consumer
(33:56):
or people like you and mein their day-to-day lives
are what we already discussed.
So we are seeing today that we get
kind of a level of qualityand speed of delivery services
that would've been unimaginablea couple of years ago.
So today, at least ifyou live in a major city,
(34:17):
you can literally go online
if you run out of toiletpaper, if you wish,
and get it delivered the same day
or, at worst case, the nextday if you really need it.
So that level of serviceand convenience is,
it's not gonna stop there,
there's gonna be furtheradvancements in this space.
So just customer utility isgoing to continue to grow
(34:40):
based on these innovationsin the logistics space,
but that's honestly not just
the only thing that we care about.
The other very big benefit
that may be a little bit lessobvious is sustainability,
and that affects every singlehuman being on this planet,
given what we've been seeingover the last couple of years
in terms of climate change and the like.
The logistics and transportation industry
(35:02):
is one of the main contributorsto carbon emissions,
and therefore to the things
that eventually cause climate change.
And if we don't manage todecarbonize this industry,
we will not be able to stopit or to even slow it down.
And so one of the reasonswhy we care deeply
about using these new sets of methods
(35:24):
to design better future logistic systems
is that, at least my personal belief
is that without these types of methods,
without these types of innovations
that we seek to develop in the new lab,
we will just not beable to meet the targets
that the industry has set itself
in terms of rapid decarbonization.
(35:45):
And the faster we candecarbonize this industry,
the better for everyone.
But with traditional methods,
there's just a limit to how much
we can squeeze out of the system,
because it is a very, veryhard to decarbonize industry
so you really need to have thesmartest possible algorithms
and models in place to makeyour existing logistics systems,
(36:09):
and especially yourfuture logistics systems,
as efficient as possible.
So that's one major kind ofdriver of value, I believe,
to anyone on this planet,
even people who may neverconsume these logistics services.
And last but not least,
I mean, we talked a lot about cities,
my other lab, the Megacity Logistics Lab,
is focused on urban logistics
(36:30):
but a large part ofthis planet's population
is still not living in cities,
and also many people on our planet
cannot afford thingslike same day delivery.
And basically making thosehigh-end logistics services
more accessible to a broaderpart of the population,
(36:52):
to other sociodemographic kindof groups within our society,
but also to people who just do not live
in major urban centers,
that is another goal,I believe, of our lab,
to basically not createkind of two separate worlds
in terms of logistics serviceand logistics innovation,
(37:12):
but ideally to make thoseinnovations accessible
to as many people as possible.
- You've mentioned thedifferent kind of things
that create value for the end consumer,
and you've told me this, correct,
that last mile delivery represents
40% of total logistic costs for companies.
So I can imagine fromthat economic standpoint
that looking at that is gonnabe a value for the company,
(37:33):
is there anything even close
that companies are dealing with
that are close to that40% of logistic cost,
or is that really the top?
- Obviously it depends a little bit
on which industry you're in,
what product you're building and selling,
but on average that's apretty good proxy, like 40%.
And I mean, think of,
I don't know, the clotheswe're wearing today,
(37:54):
they've probably beenshipped around the world
at least once, and still that last mile,
so really just getting it to you
or to the store where you bought it
from one of the big warehouses nearby.
- Well, I sew my own so I'm (indistinct).
(Matthias laughing)
All self-made, no, just-
- In that case you will haveto come up with the cost
but, in general, like that last mile,
(38:15):
that shortest littlepiece of the supply chain
is really still by farthe biggest contributor
to overall supply chain cost.
And that's also one of the reasons why,
even though problems likethe vehicle routing problem
have been tackled for decades,
it's still worth trying tosolve them in an even better way
because even small little improvements
(38:36):
in the way we execute the last mile
have tremendous impacts on our cost basis.
So, yeah, in that sense itis still one of the main
kind of also economic drivers of success
for a company that's supply chain driven.
And that eventually also boils down
(38:56):
into the cost that a consumer has to pay
to, for instance, receivea certain product.
- I imagine, also, that last mile
accounts for a pretty good portion
of the emissions that are,
vehicle emissions coming out as well.
I just hope you would speak to that a bit.
- Yeah, so it's probablyimportant to distinguish between
kind of the emissions footprint
(39:18):
that is caused by lastmile logistics directly
versus indirectly.
So if you think of last mile logistics
as that last delivery route
of one of those big brown,yellow, white vans out there,
sure they are creating emissions
but, let's say, in thegrand scheme of things
if you look at greenhouse gas emissions,
(39:39):
they might not playsuch an important role.
So electrifying, for instance,
those last mile fleets is still important,
but it will probably notnecessarily turn the needle
when it comes to avoiding orslowing down climate change.
Those last mile deliveryroutes are, however,
for instance, a major contributor
to pollutant emissions thathave immediate health impacts.
(40:01):
Think of particulate matterand NOX and the like.
So that's actually one ofthe more immediate effects
because especially in an urban environment
where you have a lot ofpeople, a lot of traffic,
a lot of these commercial vehicles
operating in a kind of close space,
those immediate health andother kind of related aspects,
those immediate healthimpacts are very measurable
(40:23):
and one major reason for companies
to seek to replace those fleets
by less polluting technologies.
But probably the moreimportant thing to consider
from a greenhouse gasemissions point of view
is actually the indirect emissions
caused by last mile logistics.
So imagine a company who starts offering
(40:44):
things like same day, onehour, on demand delivery.
Sure you have a last mile delivery route
that needs to get that productonce it's ordered from,
let's say, the store to the customer,
and that does have emissions,
but the much bigger emissions impact
is actually related to the question
how the product gets to thatstore in the first place.
Because suddenly you areoperating out of a network
(41:07):
where you're not fulfillingall your customer orders
out of a network of only a few really big,
highly consolidated warehousesor distribution centers,
but instead you are actually operating
out of a highly fragmentednetwork of stores,
micro-fulfillment centers,whatever you might call it.
So you have to movestuff around your network
(41:29):
much more frequently and ina much more fragmented way,
and that creates basicallymore miles driven per package,
if you wanna call it that way,
and therefore also muchmore emissions per delivery,
basically further upstreamyour distribution system.
It's one of the reasons whyfocusing on last mile alone
(41:51):
when optimizing a logistics process
for something likesustainability emissions
is probably missing the point.
So you really have to think about
these services at a systemic level.
So, sure, you still have tosolve that last mile problem,
but especially when it comesto emissions and the like,
you also have to considerthe entire upstream process
(42:13):
that gets the product
close enough to thecustomer in the first place
to even offer something likea one-hour delivery service.
And that's why this isso hard to decarbonize,
because some of thesetransportation activities
that happen in between
require you to overcomelarge distances by truck
(42:33):
or, worst case, by planeor something like that.
And these are just, at leastcurrently, transportation modes
and transportation lanes thatare very hard to decarbonize
from a technological point of view.
So obviously there's a lot of research,
and we are not involved inthat, that tries to make,
I don't know, planes andtrucks more sustainable,
less polluting, but at the endof the day what we care about
(42:56):
is how do we avoidunnecessary transportation
in the first place?
And that brings us back to talking about,
for instance, machine learning methods
to help us anticipatecustomer demand better,
because if we already can anticipate
that customer A is goingto order this product
within the next two hours
and he or she will want it within an hour,
(43:18):
then that puts us in a position
where we can pre-positionthe corresponding inventory,
so we can basically move the item
that he or she is going to buy
to a place close to thatcustomer already early on.
And that, again, allows us to
consolidate those transportationflows more effectively,
and therefore reducethe number of vehicles,
(43:39):
reduce the number of miles driven
to basically achieve thesame level of service.
- Is there anything you wishI was asking that I haven't?
Anything I should have been asking
that folks might wanna know about?
- A lot of people, I think,especially as we talk about
AI and machine learning,
(43:59):
are still struggling a little bit to see
the true potential of these methods.
And I'm not saying thatwe have figured it out,
I'm not saying that I can really
speak to the true potential,
but let's say one of the reasons
why I'm particularly grateful to Mecalux
for giving us this generousseed funding for the new lab
is that with them we found apartner who actually believes
(44:24):
in the significantpotential of these methods
because, let's face it, we are only
at the beginning of a journey here.
So we have to do a lot of the groundwork
on the research side first
before we will be able to reap
the full potential of these methods.
That means that maybefor the first couple of,
hopefully in a couple, but forthe first few years at least
(44:47):
of the existence of this new lab,
we will make significantadvances on the research side,
but it might take a little while
until we can turn thisinto tools and methods
that can be scaled to industrial scales,
and therefore basicallyhave an immediate ROI
for companies who wish to kind of employ
(45:07):
these types of tools and methods.
And basically finding someone
who trusts in our abilityto eventually get there
and who gives us the runway
to explore the fundamentals of this space,
build the foundation of thisnew lab in the right way
is what's really critical forus to make this lab a success.
(45:29):
So I'm particularlyexcited about this piece,
really trying to understandthe fundamental building blocks
of what will shape thefuture research of this lab
with the support of Mecalux,
and hopefully many other sponsors as well.
But at the same time, Iobviously wanna bring this
to a point where we can reallysolve real-world problems
(45:51):
at an industrial scaleas quickly as possible
because, as I said before,
the problems that we're trying to solve
aren't necessarily waiting on us.
So climate change will not wait
until we've figured out machine learning,
but it's there, it's moving rapidly
and we need to catch up quickly
in order to have a chance to fix it.
That's obviously a little bit daunting
(46:12):
but at the same time excitingto embark on that new journey,
and I'm sure with the students
and the research team that I have,
or that we all have at the center,
we are more than well-equippedto be successful at this,
so let's get started.
- Our guest today has beenDr. Matthias Winkenbach
who holds myriad titles,including Director of Research
(46:33):
at the MIT Center forTransportation and Logistics,
Director of the Megacity Logistics Lab,
as well as the CAVE Lab,
and Director of the newIntelligent Logistics Systems Lab.
Matthias, it's been apleasure having you with us.
Thanks for joining us on"Supply Chain Frontiers".
- Thank you for having me.
- Thanks for listening to this episode
of "MIT Supply ChainFrontiers", presented by the
MIT Center forTransportation and Logistics.
(46:53):
(upbeat music)
To check out other episodes of"MIT Supply Chain Frontiers",
visit ctl.mit.edu/podcasts.
And for more on the center's research,
outreach, and education initiatives,
make sure to visit usonline at ctl.mit.edu.
Our producer is Dan McCool
of the MIT Center forTransportation and Logistics.
Our sound editor is DaveLishansky of David Benjamin Sound,
(47:16):
and our audio engineertoday is Kurt Schneider
of MIT Audio Visual Services.
"MIT Supply Chain Frontiers" is recorded
on the MIT campus inCambridge, Massachusetts.
Until next time, this is Benjy Kantor
bringing you the latest insights
from the leading edge ofsupply chain management
on "MIT Supply Chain Frontiers".
(upbeat music)
- Before we get startedwith today's episode,
we have some exciting news to share.
We're pleased to welcome a new member
to the MIT Global SCALE Network,
the United Kingdom SupplyChain Excellence Center.
This newest scale center has just launched
through a partnership between
the MIT Center forTransportation and Logistics
and Loughborough University in the UK.
The UK Supply Chain Excellence Center,
based in Loughborough, willoffer a master's degree program
(00:22):
in supply chain management
based on the curriculum developed at MIT
and will soon begin admitting students
to its inaugural cohort,the class of 2026.
To learn more, visit scale.mit.edu.
(lively music)
Welcome to MIT "Supply Chain Frontiers",
presented by the Center forTransportation and Logistics.
I'm your host, Benjy Kantor.
(00:44):
Each episode of "Supply Chain Frontiers"
features center researchers and staff
or experts from industry
for in-depth conversationsabout supply chain management,
logistics, education, and beyond.
Today, our guest is Dr.Matthias Winkenbach,
a principal research scientist
and newly appointed director of research
at the MIT Center forTransportation and Logistics,
as well as the director ofthe Megacity Logistics Lab,
the Computational Analytics Visualization
(01:05):
and Education Cave Lab.
And soon, thanks tofoundational support from
and collaboration withMecalux International,
the newly launched IntelligentLogistics Systems Lab,
all of which are underthe purview of MIT CTL.
First, this episode of MIT"Supply Chain Frontiers"
is brought to you by
the MITx MicroMasters® Programin Supply Chain Management.
Gain an end-to-end understanding
(01:25):
of supply chain managementover five courses
and a final comprehensive examrepresenting the equivalent
of one semester of coursework at MIT.
Boost your skills at workor pick up some knowledge
as part of your lifelong learning.
To learn more, visitmicromasters.mit.edu/scm.
Today, vehicle routing isone of the most studied,
if not the most studiedissue in logistics,
(01:46):
stretching back about 100 years.
But despite such attention
and major strides in the decades since,
optimal efficiency in vehicle routing
and last-mile logisticsremains just out of reach.
But the comparativelysmall amount of ground
that's left to coverstands on a global scale
to have an enormousimpact on supply chains,
and AI and machinelearning might be the keys
to finding the remainingpieces of this puzzle.
(02:07):
It's not as simple as gettingfrom point A to point B,
and Mathias Winkenbach is here
as an expert in urban logistics,
network design, data analytics, and more.
And he's here today with us
to discuss his very exciting work
with using AI and machine learning
to fully optimize vehiclerouting and last-mile delivery.
Matthias, welcome to"Supply Chain Frontiers".
We're so happy to have you.
- Thanks, Benjy, for having me.
- Let's start with some context.
(02:27):
Vehicle routing is one of the most
studied problems in logistics,
but also one whose solutionis the most elusive.
Can you give us a briefoverview of, as best you can,
the last 100 years orso of vehicle routing,
or if you even can go that 100 years?
- Sure. I mean, the thing about
the vehicle routing problem,
and that's actually why wealso focus our research on this
with these new methods on the horizon,
(02:47):
is that it is one of themost well-understood,
well-studied problems inindustry but also in academia.
So if you look back,the foundational basis
of vehicle routing problems
is called the traveling salesman problem.
That's a problem that academics
have basically been studyingsince the early 1800s.
The vehicle routing problem,
which we today mostly know and study
(03:10):
in the context of logisticsand transportation,
that was first formallyintroduced in 1959, I believe,
by a researcher called Dantzig and Ramser.
So it's been a while since we'vebeen studying this problem,
and so the state of the art in this space
is also pretty advanced by now.
So there are a lot of reallygood algorithms, models,
(03:31):
to capture a lot of thereal-world complexity
that comes along with real-worldvehicle routing problems,
but the way they are solved today
still struggles with basically
getting the last, let's say, 20% right.
And those last 20% didn'treally matter that much
for a long time, when we were in a world
(03:51):
where people didn't shop online,
and transportation did have enough time
to basically get stuff from A to B,
get stuff from a warehouse to a customer
or a shop or a factory.
But in the last couple of years,
the industry has seen somuch increase in pressure
from consumer expectations.
Things had to move faster, hadto be more customer-centric,
(04:13):
had to be more individualized.
And suddenly, for these highlyresponsive logistics services
that people are asking for today,
those last 20% matter a lot.
And that's where existing methods
tend to reach the boundaryof what they're capable of.
- Now, I first heard the term last mile
with regard to people moving, commuting.
(04:35):
I worked for and I was a member
of a public bike-share system,
and it was around 2014 or 15 or so.
Is that the kind of thingyou're looking at also,
or is it specificallycommercial, industrial,
the things like Amazon gettingyour package to your door
as opposed to getting you to work on time?
(04:57):
- I mean, there are certain similarities,
but what makes last-mile logistics,
so basically the commercialmovement of goods
on that last mile, particularly complex
is that we are dealing withwhat we call fragmented routes.
So if you are a commuter,
you're usually an individual
who has to get from point A to point B.
(05:17):
And yes, you're making routingdecisions along the way,
but in a way you're just trying to find
the shortest or mostconvenient or fastest path
from point A to point B.
While someone likeAmazon, UPS, you name it,
they have to deliver to multiple customers
along one consolidatedand efficient route,
so think of the big brownvan on the road out there.
(05:40):
That van typically makesaround 120 stops a day,
and those have to be sequenced in a way
that maybe you're not alwaysable to take the shortest path
between any two customersalong that route,
but your goal is tooptimize the overall route.
So to, for instance,find the overall fastest,
cheapest, shortest paththat connects all of these,
(06:03):
in this case, 120 pointson the map, basically.
- And when we're talking about
the full efficiency of that last mile,
and when I say efficiency Imean sort of all these things,
like the cost and the speed
and whatever the other factors are.
Why has it been so elusive to get to that,
the thing that makes it work completely?
(06:25):
- There's a few thingsthat make this really hard.
One is it's not just a mathproblem; it's a human problem
because one major source of uncertainty
in routing is, for instance, the driver.
Whether the driver isvery experienced or not,
whether the driver is having a good day
(06:46):
or a bad day that morning
is also a factor of how other participants
in a mobility system likea city, for instance,
interact with that one route.
So, for example, companies like UPS
can optimize the heck out of a route,
but then what happens if unexpectedly
(07:07):
there's a traffic accident?
And that completelychanges your assumptions
around travel speeds,
travel times from onecustomer to the other.
So there's a lot of uncertainty,
a lot of factors thatare very hard to control,
if not impossible tocontrol for the company
when it tries to findthe optimal solution.
And that's where, let'ssay, traditional solution
(07:28):
approaches to thisproblem tend to struggle,
because they're eithernot capable of capturing
all those uncertain influences,
or they're not kind of efficient enough
to quickly respond to changesin the operational environment
to basically find the next best solution.
- And I know you've mentioned in the past
that the complexities around this
(07:50):
are based a lot in density.
Now, if we're talking about density
we're talking about population,
are we talking about infrastructure?
What are those things thatare standing in the way
that are challenging,
or that create theproblems in the first place
that aren't necessarily standingin the way but they exist,
and so you have to figure it out?
- So, I mean, there's a reason why
that one lab that I've been leading
(08:11):
for the last 10 years roughly
is called the Megacity Logistics Lab,
Because when we startedout doing this research
we very much focusedon really large cities.
Why? Because that's wherea lot of people live
But, also, that's wherelogistics processes
are the most complex.
And one of the main drivers
of that complexity is indeed density,
So every city, or let'ssay most cities out there
(08:34):
weren't planned for the size
and the amount of peopleand the amount of goods
that need to move into andout of these cities today.
Most cities out there grewhistorically, and that's okay
but the problem is that usually
transportation infrastructuredoesn't grow proportionally,
because it was never designed
to really be able to dealwith that much traffic,
(08:57):
that much demand for this limited space
on the road, basically.
And that's where, as cities grow,
as we see ongoing urbanizationall around the world,
density in these cities increases
and that means traffic density increases,
demand density increases.
And that, again, has bothpositive and also negative effects
(09:18):
on logistics processes.
The positive thing is,
if your customers aremore densely co-located,
then at least theoreticallyyou have to travel less
to meet the same number of people
and to deliver the sameamount of goods, for instance.
So technically, your delivery routes
should get shorter, faster, cheaper.
But the downside of itis, as I said before,
transportation infrastructuredidn't grow proportionally
(09:40):
so you have more traffic,more uncertainty,
generally mobility decreases,travel speeds decrease,
so that makes these routesnot just less efficient
but also less predictable
because there's a higher likelihood
of a random incidentlike a traffic accident
in a high-density environment
as compared to, let's say,a low-density environment
like rural Massachusetts.
(10:03):
- I'm glad you brought upthe Megacity Logistics Lab.
I'm kind of curious, can you tell us
how that lab fits into whatthe current methods are,
what the new methods are,considering it's only 10 years,
it's only been around for 10% of the time
that this problem has been focused on.
And so what are the new methods
for looking at these older problems,
(10:24):
and how do they differ from
what people were trying to do before?
- Well, I mean, when I startedresearching this space,
which was shortly beforethe Megacity Lab started,
but that was basically at the onset
of widespread e-commerce.
That's when Amazon juststarted operating, basically.
That's when people kind of did realize
(10:44):
how important the last mileof any logistics process
might become in the future,
but probably nobody reallyhad it on their radar
that today we're talking about services
like order now and get itwithin the next two hours
or something like that.
And so the traditional methods
that also my labpredominantly has been using
(11:07):
over the last 10 years
can be generally classified
as what we call operationsresearch methods.
So this, broadly speaking,means optimization methods.
And here we mostly distinguishbetween exact methods,
so these are typically mathematical models
where you formulate, let'ssay, a routing problem exactly.
(11:27):
So you write it down as a set of equations
and then you're using algorithms
to solve those equations exactly.
So, ideally, at the end of the day
you want to get the optimal solution.
The problem about these methods is
that as soon as you applythem to real-world problems
that are of a real-world size,
like for instance those 120stops of the big brown van
(11:48):
that I mentioned earlier,
those methods take a very long time
to find that optimal solution,if they can find it at all.
So they're not efficient enough
to really be applied at scale, typically.
That's why the other family of methods
that fall under kind oftraditional operations research
would be heuristics andmeta heuristics approaches,
(12:08):
which is basically a set of rules,
a set of kind of decision procedures
that you put into an algorithmthat are often tailored
towards the specifics of the problem.
So for instance, in thecontext of vehicle routing,
you can encode certain rules
that define what is agood or a not so good move
(12:32):
of a vehicle from one customer to the next
based on your knowledge of the problem,
based on your domain expertise.
So these methods tend to be faster,
tend to be more efficient infinding really good solutions.
The downside is that theytypically can never guarantee
that they find the optimal solution.
(12:53):
So that's just kind ofa two-minute summary
of what those traditional methods are.
And my lab has been veryactive in publishing papers,
working with industry on using
both of these types of methods extensively
on solving all sorts oflast-mile logistics problems,
not just routing.
So the reason why we are nowexploring a new set of methods
(13:19):
inspired by all this generative AI
and machine learning stuffthat's going on out there today
is that, as I said before,the world is moving faster
than these traditionalmethods can keep up.
So for instance, being able tooptimize the logistics system
for services likeon-demand one-hour delivery
(13:39):
means you are optimizing a route
with very little time at hand.
And then some of these existing methods,
some of these traditional methods
might just not be fast or efficient enough
to really solve your problemwithin that little time.
Similarly, the trend towardsmore customer centricity,
so really focusing logistics processes
(13:59):
around the individual requests,
the individual needs ofindividual customers,
exposes them to so much uncertainty
that, again, these traditional methods
have a hard time capturing all of that,
capturing the uncertainty andthe real-world complexity of,
let's say, a modern-agestate-of-the-art routing problem.
And that's where we need approaches
that are more data-driven,
(14:21):
that learn more fromreal-world operational data,
that learn more aboutthe real characteristics
of individual customersor individual drivers,
or individual routes.
And these data-drivenmethods can't really be found
in the traditionaloperations research space,
but are very prevalent in whatis currently being discussed
as machine learning and AI methods.
(14:43):
- I'm really glad youbrought up that idea of needs
and how efficiently someone might need
to have a package delivered to them,
or how quickly a companyneeds to have delivered.
And I'm gonna come up witha more challenging question
for later in the show,
but something else that you touched on is
as you are working more and more with AI,
machine learning, andpredictive analytics,
how is that AI differ from simply someone
(15:06):
or a group of people analyzing data
that they already know exists,
and making a prediction based on that?
- That's the thing, Imean, there's been research
on predictive analytics for decades.
And I'm also not sayingthat this will become
irrelevant in the future,
but to give you a very tangible example,
(15:27):
if you're looking atdemand forecasting today,
there are very capablemethods to forecast,
let's say, the demandfor a certain product in,
let's just say, Amazon's assortment
at the scale of an entire city
for maybe the next day or the next week.
So you can probablyrelatively accurately predict
(15:49):
how frequently that product will be bought
in the city of Boston next week.
But if we're talking about
customer-centric logistics processes
with very short lead time,
let's stick to the example ofon-demand one-hour delivery.
Being able to predict how often
that product will be bought next week
is pretty much useless for you,
what you really need isthe ability to predict
(16:12):
at a very high spatial butalso temporal resolution
when and where that productis going to be bought.
So basically, what youneed to be able to do
is to have a good estimate
of how many items of thatproduct will be bought,
let's say, in a zip code in Cambridge
(16:32):
within the next 30 minutes.
And that's where those traditional methods
that rely on a lot of data tomake kind of point estimates
or distribution estimatesare not that useful anymore,
because you need tools thattake much more data into account
but also detect muchmore interdependencies
between the various data points
(16:54):
that you use to train these models,
to be able to make thesemore fine-grain predictions.
That's one difference between
what used to be just prediction
versus what we now call
machine learning-basedpredictive analytics.
But the other thing thatI would like to point out,
and that brings us back to routing, is,
for instance, up until recently,
(17:15):
predicting a really good vehicle route
was not something that wasconsidered really feasible
because, let's say, if you are to predict
a good solution to a routing problem,
you're not just makingone point prediction,
you're not just predicting onevalue far out in the future,
you're basically predicting a solution
(17:36):
to a sequential decision problem.
So you're predicting asequence of stops of a vehicle
that has to make sense.
Not just one stop that has to make sense,
the whole sequence ofstops has to make sense,
it has to be a solutionthat's both feasible,
that's even viable to execute on,
and at the same time should be
(17:56):
as close to an optimalsolution as possible.
And that's something where,
let's say, traditional prediction methods
were just not capable of doing that.
Now, with the advent of stufflike transformer models,
so actually more or lessthe same kind of methods
that also power things like ChatGPT
and all of these chatbots,
even though they areusing these methods in,
(18:17):
let's say, the context ofnatural language processing,
these types of methodsalso have a huge potential
for optimization problemsin the logistics industry,
like the vehicle routing problem.
- Something funny that cameup recently in a conversation
I was having with Jose Velasquez,
with regard to some Amazondelivery and last-mile stuff
was that some of the thingsthat seem very intuitive,
(18:38):
for instance, that you mightthink that it is most efficient
to deliver packages in a way
that it has the best geographical routing,
but actually that mightnot be the most efficient
in terms of sustainability,and even speed,
because maybe it has to dowith how heavy the packages are
and things like that.
- So another example in that space is,
and something that we alsoexplored a few years ago already
(19:01):
with folks like Amazon and others,
is you wanna come up with route plans
that are not just short or cheap or fast,
but that a driver can actually execute on
and that a driver actuallythinks is a good route.
And here's one of the big benefits
of these more data-drivenlearning-based approaches
(19:22):
as compared to just runningan optimization model,
because a driver, especiallyan experienced one,
knows exactly how to operate a route.
And if you look at manydrivers out there today,
they get a route plan in the morning
and they make their own adjustments
because, for instance,they know much better
than any routing algorithmswhere to park safely,
(19:43):
where to basically avoid traffic
at certain times of the day,
or they might even know which customer
might be more or less available
during which times of the day,
even though that customer may have never
explicitly voiced a preference for,
let's say, being deliveredbetween 10 and 11 AM,
but the driver knows this.
(20:05):
And that kind of consistency,
that kind of tacit knowledge
that's ingrained in every single driver,
is basically impossible to encode
in a hard optimization algorithm
or a set of mathematical equations.
But it can be learned from observing
the actual behavior of drivers over time,
(20:26):
so we may not necessarilycome up with shorter
or faster or cheaper routes,
but we may come up withroutes that are more reliable
because the driver is actually able to
act on the route plan that hegets in the morning, or she,
and also some of hisor her tacit knowledge
about that route is alreadyembedded in the plan.
(20:48):
- And is that gonna changeor has it been changing
the more there is autonomous delivery,
or drone delivery or things like that?
- I mean, today, for better or for worse,
there is not that muchautonomous delivery yet,
especially not in urban contexts
but let's say if we wereever to get to that point,
what you wouldn't wanna lose
if you were fully automatingthe delivery process
(21:10):
is that tacit knowledge about,
for instance, individual customers
or safe and availableparking spaces and the like.
And so you would wanna have
some sort of trained modelrunning in the background
that tells that autonomous vehicle,
or drone or whatever it might be,
how to best deliver theproduct to your customer,
(21:31):
even though there is nohuman involved anymore.
So in a way, if we want these systems
to be as close to a humandelivery experience as possible,
then sure, this is the kind of method
that we should explore further.
- And when we're talkingabout route optimization,
have you seen trends thatcompanies are prioritizing
when they're optimizing their routes?
(21:52):
Have you seen those shift from the past?
Are there things thatyour teams are working on
that you've seen start to be implemented,
and how is that changing the landscape?
- I mean, I think that theunderlying routing problem,
I mean, as I said, that hasbeen studied for decades.
The fundamentals of that problemdon't necessarily change,
(22:14):
but what has definitely been a trend is
how can we focus our solutionsto the routing problem
more around the specificneeds of individual customers?
So, while a few years ago youtreated every single customer
more or less as a uniform mass,
now it's really moreabout individual requests
(22:36):
and individual characteristics.
And that obviously also has implications
for the way companiesapproach these problems.
They have to capture morestop-specific information,
for instance, how muchdoes it actually take me
to deliver to customerA versus customer B?
And that may depend on thenature of that customer,
(22:56):
the nature of the buildingthat the customer is in,
and many other factors.
But, also, we need to understand
preferences more explicitly.
And a few years ago peoplewould opt for something
like time window delivery,
and you would get immediateinput from your customer
as to when they wanna be delivered to.
(23:16):
Nowadays, the expectation is more like,
"Well, the logistics processshould already anticipate
my availability, my preferences."
So even though I don't voicethese preferences explicitly,
your routing algorithm should probably
already take into account
what we have learned aboutthat customer over time,
and therefore basicallymake that customer happier
(23:37):
by, under the hood, alreadycustomizing the delivery
to his or her preferences.
So that shift towardsmore customer centricity
comes in various shapes and forms
but has definitely been
the most important driver of complexity,
but also innovation in the routing space.
- I'll bring in my morechallenging question here,
which is talking aboutcustomer preferences.
(23:59):
If I order a soccer ball online,
my preference would be
that my soccer ball'sin my hand right now.
How do, I guess, we balance the,
there's one part that'sjust sort of like...
Again, the logistics of gettingthat into somebody's hand
as fast as they want it versusas fast as they need it,
(24:22):
I guess, is the question,
which is like theprioritization when it comes to-
You know there's labs at CTL
that deal with sustainabilityand humanitarian,
and if delivering that soccer ball
is gonna impede my delivering water
or baby formula or something like that.
Is that sort of, for lackof a better term, morality
in decision includedin how machines learn,
(24:44):
how people look at it, howcompanies put that into effect?
- Let's say it could be,but it's something that,
let's say, we haven'texplored the morality
of vehicle routing problem explicitly yet
because that's honestlya rather subjective term.
It comes down to the same problem
that many of those autonomous
(25:05):
vehicle manufacturers have today,
like, how do we teach a vehicle morality
if it has a choice between, I don't know,
having an accident that,let's say, kills a child
versus kills an old lady?
What's better, what'sworse? We don't know.
And as humans, if there's a human driver,
we usually don't really have a choice
because it just happens so fast.
(25:27):
A machine might actually have a choice,
so how do we teach it ethical behavior?
And that's something thatwe are probably not equipped
to do in our lab, and that's why we try to
stay away from that.
But what you can do,obviously, as a company is
you can prioritize certain objectives
in your decision-making.
So for instance, if it's maybe
(25:48):
not a life-and-death decision,
but if it's about reducing youroverall emissions footprint
as a whole, as a company, let's say,
if an e-commerce platformwants to make sure
that it still meets thecustomer requirements
as good as it possibly can,
but at the same time tries tobe as sustainable as possible,
then that may be part of that
(26:09):
rather complex set of objectives
that you can "teach"to a routing algorithm
so that that algorithm probably already
makes that distinction ona product level and say,
"Okay, if I have todeprioritize certain products
because either I run out of capacity
or because I'm hitting myemissions threshold, basically,
(26:31):
then which one should I deprioritize,
the soccer ball or the baby formula?"
And so in a way, that'sanother opportunity, I believe,
where these algorithms alsohave a unique strength,
which is learning from customer feedback.
If you get your soccer ball afew hours or maybe a day late,
(26:52):
you might not be thrilled about it
but it's also not the end of the world.
If the mother of a newbornchild runs out of baby formula
because the delivery is getting late,
she will probably be much less thrilled.
And if you can capturethat feedback somehow
and basically use that as your input,
from which you can thenlearn or train an algorithm
(27:14):
to kind of check itspriorities a little bit,
then next time you have that problem
that algorithm is already better equipped
to anticipate the kind of discomfort
or disutility that it is causing
by deprioritizing the wrong thing.
So in that sense, rather than you and I
(27:35):
who design an algorithm havingto make an explicit choice,
whether it's the formulaor the soccer ball,
we can actually havethose algorithms learn
from the response of theentire kind of social system,
which is, in this case, yourcustomers, to your priorities.
And over time, those wouldthen hopefully match.
(27:55):
So what your algorithm does
matches both your corporate objectives,
as well as the general norm
or the general prioritiesof your customer base.
- Yeah, I appreciate that.
I don't know that thequestion has an answer,
just kind of curious when talking about
the different sort of research aspects,
especially within CTL,
just thinking aboutthose kind of priorities
(28:18):
and what kind of goes into this.
Shifting gears a little bit,
I would love to hear youtalk about this new venture,
this Intelligent Logistics Systems Lab,
where the research goes from here.
Is this a step fartherfrom what the other labs
that you're directing are doing,
or is it a completelydifferent type of research?
- I mean, the formation of the
Intelligent Logistics Systems Lab
(28:39):
is basically born out of anobservation that we've had
for several years now,
namely that more and more of the research
that my other labs at CTL, butalso other labs at the center
and other places at MIT, are doing,
more and more of that work
moves away from basicallythose traditional methods
(29:01):
that we discussed before,
those traditionaloperations research methods,
and move further towardsthis "new family of methods"
in the machine learning and AI space,
and most importantly we see itas a huge potential for work
kind of at the intersection of the two,
basically hybrid methodsthat basically try to combine
(29:22):
the best of both worlds, thebest of traditional OR methods
and the best that those newML and AI methods can offer.
Because what we see in ourearly research already today
is that actually the greatestpotential for improvement
over what we've been doingfor the last couple of decades
doesn't necessarily lie in
completely replacing those methods,
(29:43):
but rather kind of putting them
on machine-learning steroids,if you allow that term.
And that's kind of anexciting observation for us
and also kind of goes beyondthe original scope of,
for instance, the Megacity Logistics Lab
and many other labs at the center.
So I think it's the righttime to actually have
(30:04):
a dedicated group at the center
that bundles our activities
in that space a little bit more,
and really envisions kindof the future generation
of logistics systems thatenable logistics services
that we probably todaycan just kind of imagine,
but that will have to become a reality
(30:25):
within the next couple of years,
because that trend towardsfurther customer centricity,
increasing pressure of sustainability,
increasing needs forsupply chain resilience,
all of these trendsrequire us to build systems
that are smarter in the sensethat they are more agile
(30:45):
in responding to changes inthe operational environment,
in the socioeconomic environment,
in the geopolitical environment,
and that are more data-driven.
That are basically, as I said before,
learn from the very rich knowledge
that exists in many individuals out there
in the logistics industry,
be that the driver, bethat the warehouse worker,
(31:05):
be that the supply chainlogistics executive
with 40 years of industry experience.
We don't wanna lose the human expertise
in designing those future systems,
but we need to be ableto scale that expertise
and basically ingrain it
in the mathematical models that we build,
(31:27):
such that we can solvethese very challenging
future logistics systems design problems
at scale and rapidly.
- This is something that I've asked
a few of your colleagues about their labs.
Is there an end goal for the lab?
Is there a point at which youhave done what you need to do,
you've antiquated the lab in a good way
(31:49):
because the solution has been found?
- Yeah, I mean, in a waythe Megacity Logistics Lab
might be a good example for that
because it's been around fora bit more than 10 years now.
We've done a lot of great work
but the lab has to someextent outgrown itself,
so we no longer just focuson last mile logistics
(32:10):
in the very narrow sense,
we are basically alreadyworking on a lot of stuff
that involves more thanjust the last mile,
because today the way youoperate a last mile vehicle route
isn't a problem that is isolated
from everything else in the supply chain,
it's basically highly interconnected
with aspects of networkdesign, inventory, planning,
(32:33):
sourcing decisions, long-haultransportation decisions.
So supply chains and logistic systems are,
or have become much more interconnected
so, basically, focusingon just that one aspect
of last mile logistics may nolonger make that much sense,
at least in many cases.
Which is one of the reasons why,
(32:54):
I mean, the lab will still exist
but I think our focus will probably shift
towards looking at supplychains and logistics systems
at a slightly higher level,
which is in part enabledby also looking at
a broader set of methods
to basically not just optimizethat single vehicle route,
(33:14):
but ideally to find an optimalsolution to an entire system
that may consist of manydifferent vehicle routes,
1,000s of customers, large networks,
many interconnected decisions.
And ideally we want to find ways
to solve these problemsin an integrated way,
so just focusing on last mile
(33:35):
in this case might notnecessarily be what we need to do
to design the next generationof logistics systems.
- We talked a little bit earlier
about sort of the satisfactionfor an end consumer.
What are the implicationsfor everyday consumers?
Again, we talked a little bit
like delivery of the soccer ball, but...
- I mean, the obvious benefit
to the, let's say, the end consumer
(33:56):
or people like you and mein their day-to-day lives
are what we already discussed.
So we are seeing today that we get
kind of a level of qualityand speed of delivery services
that would've been unimaginablea couple of years ago.
So today, at least ifyou live in a major city,
(34:17):
you can literally go online
if you run out of toiletpaper, if you wish,
and get it delivered the same day
or, at worst case, the nextday if you really need it.
So that level of serviceand convenience is,
it's not gonna stop there,
there's gonna be furtheradvancements in this space.
So just customer utility isgoing to continue to grow
(34:40):
based on these innovationsin the logistics space,
but that's honestly not just
the only thing that we care about.
The other very big benefit
that may be a little bit lessobvious is sustainability,
and that affects every singlehuman being on this planet,
given what we've been seeingover the last couple of years
in terms of climate change and the like.
The logistics and transportation industry
(35:02):
is one of the main contributorsto carbon emissions,
and therefore to the things
that eventually cause climate change.
And if we don't manage todecarbonize this industry,
we will not be able to stopit or to even slow it down.
And so one of the reasonswhy we care deeply
about using these new sets of methods
(35:24):
to design better future logistic systems
is that, at least my personal belief
is that without these types of methods,
without these types of innovations
that we seek to develop in the new lab,
we will just not beable to meet the targets
that the industry has set itself
in terms of rapid decarbonization.
(35:45):
And the faster we candecarbonize this industry,
the better for everyone.
But with traditional methods,
there's just a limit to how much
we can squeeze out of the system,
because it is a very, veryhard to decarbonize industry
so you really need to have thesmartest possible algorithms
and models in place to makeyour existing logistics systems,
(36:09):
and especially yourfuture logistics systems,
as efficient as possible.
So that's one major kind ofdriver of value, I believe,
to anyone on this planet,
even people who may neverconsume these logistics services.
And last but not least,
I mean, we talked a lot about cities,
my other lab, the Megacity Logistics Lab,
is focused on urban logistics
(36:30):
but a large part ofthis planet's population
is still not living in cities,
and also many people on our planet
cannot afford thingslike same day delivery.
And basically making thosehigh-end logistics services
more accessible to a broaderpart of the population,
(36:52):
to other sociodemographic kindof groups within our society,
but also to people who just do not live
in major urban centers,
that is another goal,I believe, of our lab,
to basically not createkind of two separate worlds
in terms of logistics serviceand logistics innovation,
(37:12):
but ideally to make thoseinnovations accessible
to as many people as possible.
- You've mentioned thedifferent kind of things
that create value for the end consumer,
and you've told me this, correct,
that last mile delivery represents
40% of total logistic costs for companies.
So I can imagine fromthat economic standpoint
that looking at that is gonnabe a value for the company,
(37:33):
is there anything even close
that companies are dealing with
that are close to that40% of logistic cost,
or is that really the top?
- Obviously it depends a little bit
on which industry you're in,
what product you're building and selling,
but on average that's apretty good proxy, like 40%.
And I mean, think of,
I don't know, the clotheswe're wearing today,
(37:54):
they've probably beenshipped around the world
at least once, and still that last mile,
so really just getting it to you
or to the store where you bought it
from one of the big warehouses nearby.
- Well, I sew my own so I'm (indistinct).
(Matthias laughing)
All self-made, no, just-
- In that case you will haveto come up with the cost
but, in general, like that last mile,
(38:15):
that shortest littlepiece of the supply chain
is really still by farthe biggest contributor
to overall supply chain cost.
And that's also one of the reasons why,
even though problems likethe vehicle routing problem
have been tackled for decades,
it's still worth trying tosolve them in an even better way
because even small little improvements
(38:36):
in the way we execute the last mile
have tremendous impacts on our cost basis.
So, yeah, in that sense itis still one of the main
kind of also economic drivers of success
for a company that's supply chain driven.
And that eventually also boils down
(38:56):
into the cost that a consumer has to pay
to, for instance, receivea certain product.
- I imagine, also, that last mile
accounts for a pretty good portion
of the emissions that are,
vehicle emissions coming out as well.
I just hope you would speak to that a bit.
- Yeah, so it's probablyimportant to distinguish between
kind of the emissions footprint
(39:18):
that is caused by lastmile logistics directly
versus indirectly.
So if you think of last mile logistics
as that last delivery route
of one of those big brown,yellow, white vans out there,
sure they are creating emissions
but, let's say, in thegrand scheme of things
if you look at greenhouse gas emissions,
(39:39):
they might not playsuch an important role.
So electrifying, for instance,
those last mile fleets is still important,
but it will probably notnecessarily turn the needle
when it comes to avoiding orslowing down climate change.
Those last mile deliveryroutes are, however,
for instance, a major contributor
to pollutant emissions thathave immediate health impacts.
(40:01):
Think of particulate matterand NOX and the like.
So that's actually one ofthe more immediate effects
because especially in an urban environment
where you have a lot ofpeople, a lot of traffic,
a lot of these commercial vehicles
operating in a kind of close space,
those immediate health andother kind of related aspects,
those immediate healthimpacts are very measurable
(40:23):
and one major reason for companies
to seek to replace those fleets
by less polluting technologies.
But probably the moreimportant thing to consider
from a greenhouse gasemissions point of view
is actually the indirect emissions
caused by last mile logistics.
So imagine a company who starts offering
(40:44):
things like same day, onehour, on demand delivery.
Sure you have a last mile delivery route
that needs to get that productonce it's ordered from,
let's say, the store to the customer,
and that does have emissions,
but the much bigger emissions impact
is actually related to the question
how the product gets to thatstore in the first place.
Because suddenly you areoperating out of a network
(41:07):
where you're not fulfillingall your customer orders
out of a network of only a few really big,
highly consolidated warehousesor distribution centers,
but instead you are actually operating
out of a highly fragmentednetwork of stores,
micro-fulfillment centers,whatever you might call it.
So you have to movestuff around your network
(41:29):
much more frequently and ina much more fragmented way,
and that creates basicallymore miles driven per package,
if you wanna call it that way,
and therefore also muchmore emissions per delivery,
basically further upstreamyour distribution system.
It's one of the reasons whyfocusing on last mile alone
(41:51):
when optimizing a logistics process
for something likesustainability emissions
is probably missing the point.
So you really have to think about
these services at a systemic level.
So, sure, you still have tosolve that last mile problem,
but especially when it comesto emissions and the like,
you also have to considerthe entire upstream process
(42:13):
that gets the product
close enough to thecustomer in the first place
to even offer something likea one-hour delivery service.
And that's why this isso hard to decarbonize,
because some of thesetransportation activities
that happen in between
require you to overcomelarge distances by truck
(42:33):
or, worst case, by planeor something like that.
And these are just, at leastcurrently, transportation modes
and transportation lanes thatare very hard to decarbonize
from a technological point of view.
So obviously there's a lot of research,
and we are not involved inthat, that tries to make,
I don't know, planes andtrucks more sustainable,
less polluting, but at the endof the day what we care about
(42:56):
is how do we avoidunnecessary transportation
in the first place?
And that brings us back to talking about,
for instance, machine learning methods
to help us anticipatecustomer demand better,
because if we already can anticipate
that customer A is goingto order this product
within the next two hours
and he or she will want it within an hour,
(43:18):
then that puts us in a position
where we can pre-positionthe corresponding inventory,
so we can basically move the item
that he or she is going to buy
to a place close to thatcustomer already early on.
And that, again, allows us to
consolidate those transportationflows more effectively,
and therefore reducethe number of vehicles,
(43:39):
reduce the number of miles driven
to basically achieve thesame level of service.
- Is there anything you wishI was asking that I haven't?
Anything I should have been asking
that folks might wanna know about?
- A lot of people, I think,especially as we talk about
AI and machine learning,
(43:59):
are still struggling a little bit to see
the true potential of these methods.
And I'm not saying thatwe have figured it out,
I'm not saying that I can really
speak to the true potential,
but let's say one of the reasons
why I'm particularly grateful to Mecalux
for giving us this generousseed funding for the new lab
is that with them we found apartner who actually believes
(44:24):
in the significantpotential of these methods
because, let's face it, we are only
at the beginning of a journey here.
So we have to do a lot of the groundwork
on the research side first
before we will be able to reap
the full potential of these methods.
That means that maybefor the first couple of,
hopefully in a couple, but forthe first few years at least
(44:47):
of the existence of this new lab,
we will make significantadvances on the research side,
but it might take a little while
until we can turn thisinto tools and methods
that can be scaled to industrial scales,
and therefore basicallyhave an immediate ROI
for companies who wish to kind of employ
(45:07):
these types of tools and methods.
And basically finding someone
who trusts in our abilityto eventually get there
and who gives us the runway
to explore the fundamentals of this space,
build the foundation of thisnew lab in the right way
is what's really critical forus to make this lab a success.
(45:29):
So I'm particularlyexcited about this piece,
really trying to understandthe fundamental building blocks
of what will shape thefuture research of this lab
with the support of Mecalux,
and hopefully many other sponsors as well.
But at the same time, Iobviously wanna bring this
to a point where we can reallysolve real-world problems
(45:51):
at an industrial scaleas quickly as possible
because, as I said before,
the problems that we're trying to solve
aren't necessarily waiting on us.
So climate change will not wait
until we've figured out machine learning,
but it's there, it's moving rapidly
and we need to catch up quickly
in order to have a chance to fix it.
That's obviously a little bit daunting
(46:12):
but at the same time excitingto embark on that new journey,
and I'm sure with the students
and the research team that I have,
or that we all have at the center,
we are more than well-equippedto be successful at this,
so let's get started.
- Our guest today has beenDr. Matthias Winkenbach
who holds myriad titles,including Director of Research
(46:33):
at the MIT Center forTransportation and Logistics,
Director of the Megacity Logistics Lab,
as well as the CAVE Lab,
and Director of the newIntelligent Logistics Systems Lab.
Matthias, it's been apleasure having you with us.
Thanks for joining us on"Supply Chain Frontiers".
- Thank you for having me.
- Thanks for listening to this episode
of "MIT Supply ChainFrontiers", presented by the
MIT Center forTransportation and Logistics.
(46:53):
(upbeat music)
To check out other episodes of"MIT Supply Chain Frontiers",
visit ctl.mit.edu/podcasts.
And for more on the center's research,
outreach, and education initiatives,
make sure to visit usonline at ctl.mit.edu.
Our producer is Dan McCool
of the MIT Center forTransportation and Logistics.
Our sound editor is DaveLishansky of David Benjamin Sound,
(47:16):
and our audio engineertoday is Kurt Schneider
of MIT Audio Visual Services.
"MIT Supply Chain Frontiers" is recorded
on the MIT campus inCambridge, Massachusetts.
Until next time, this is Benjy Kantor
bringing you the latest insights
from the leading edge ofsupply chain management
on "MIT Supply Chain Frontiers".
(upbeat music)
Popular Podcasts
Amy Robach & T.J. Holmes present: Aubrey O’Day, Covering the Diddy Trial
Introducing… Aubrey O’Day Diddy’s former protege, television personality, platinum selling music artist, Danity Kane alum Aubrey O’Day joins veteran journalists Amy Robach and TJ Holmes to provide a unique perspective on the trial that has captivated the attention of the nation. Join them throughout the trial as they discuss, debate, and dissect every detail, every aspect of the proceedings. Aubrey will offer her opinions and expertise, as only she is qualified to do given her first-hand knowledge. From her days on Making the Band, as she emerged as the breakout star, the truth of the situation would be the opposite of the glitz and glamour. Listen throughout every minute of the trial, for this exclusive coverage. Amy Robach and TJ Holmes present Aubrey O’Day, Covering the Diddy Trial, an iHeartRadio podcast.
Betrayal: Season 4
Karoline Borega married a man of honor – a respected Colorado Springs Police officer. She knew there would be sacrifices to accommodate her husband’s career. But she had no idea that he was using his badge to fool everyone. This season, we expose a man who swore two sacred oaths—one to his badge, one to his bride—and broke them both. We follow Karoline as she questions everything she thought she knew about her partner of over 20 years. And make sure to check out Seasons 1-3 of Betrayal, along with Betrayal Weekly Season 1.
Crime Junkie
Does hearing about a true crime case always leave you scouring the internet for the truth behind the story? Dive into your next mystery with Crime Junkie. Every Monday, join your host Ashley Flowers as she unravels all the details of infamous and underreported true crime cases with her best friend Brit Prawat. From cold cases to missing persons and heroes in our community who seek justice, Crime Junkie is your destination for theories and stories you won’t hear anywhere else. Whether you're a seasoned true crime enthusiast or new to the genre, you'll find yourself on the edge of your seat awaiting a new episode every Monday. If you can never get enough true crime... Congratulations, you’ve found your people. Follow to join a community of Crime Junkies! Crime Junkie is presented by audiochuck Media Company.