We sit down with world-renowned supply chain guru, Professor David Simchi-Levi. This MIT professor and head of the MIT Data Science Lab, shares his insights on supply chain digitization and delivering on the promise of increased forecast accuracy while reducing risks. It's a transformative discussion on how you, with a moderate financial investment, can swiftly digitize your supply chain while reaping the rewards. Our discussion touches on:
- How a unified view of demand and segmentation can allow you to craft bespoke supply chain strategies for different products, channels, and markets.
- The fusion of analytics, data, and algorithms with intuition, experience, and domain knowledge.
- The pioneering work done by the MIT Data Science Lab on supply chain resiliency and how giants like Ford Motor Company leveraged this technology.
We're not just talking theories. We dive straight into how supply chain management can be optimized using smart S&OP and smart execution. Learn We share real-life examples of how a consumer-packaged goods manufacturing company used data sources to predict retail orders, making concepts easy to grasp and apply in your own businesses.
In this discussion we reference:
- Harvard Business Review article by David Simchi-Levi, A Simpler Way to Modernize Your Supply Chain
- The MIT Data Science Lab
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Speaker 1 (00:01):
Before the pandemic, the perception that executive
had about supply chaindigitization was that it
required significant financialinvestment.
It will take many years peoplewere talking about five to six
years to digitize the entiresupply chain, because it
(00:23):
required instrumentation ofevery process and every product
and every facilities.
What we learned during thepandemic is that the future is
here, that with moderatefinancial investment, by taking
advantage of existing data thatcompanies have and complementing
(00:47):
it with additional data thatcompanies can get from
third-party partners, we canachieve almost all the benefits
of full-flight supply chaindigitization in a relatively
short period of time.
Speaker 2 (01:04):
From LAI.
This is Shelf Life.
What is supply chaindigitization and why does it
(01:25):
matter now?
How can you better use the datayou already have to increase
your forecast accuracy?
How can you reduce supply chainrisks in a time of constant
disruptions?
On every episode of Shelf Life,we answer questions like these
and more, with the help ofleaders across the consumer
goods industry.
Today, I'm very happy towelcome Professor David Simche
(01:47):
Levy, Professor of EngineeringSystems at MIT and head of the
MIT Data Science Lab.
David is widely considered oneof the leading voices in supply
chain management, havingpublished four books on the
subject and countless articlesDuring his career.
He's also founded severalcompanies focused on analytics
and supply chain optimization.
I'm your host, Joel Beale, CEOof Aloe AI.
(02:10):
We'll be back with David rightafter this.
Speaker 3 (02:29):
There's plenty of data to help, but pulling and
analyzing reports from retailerportals is so tedious and time
consuming that you can't respondto problems at the shelf until
the bull whip hits you in theface.
That's cow, actually.
Now there's Aloeai to help.
Aloe automatically aggregatesand harmonizes data from your
retailers, supply chain partnersand ERP.
(02:50):
Then we make it easy to findinsights using pre-built
consumer goods, specific metricsand dashboards, so you can
sense, predict and respondinstantly.
Check it out today and get ademo at Aloeai.
Speaker 2 (03:04):
David, welcome to Shelf Life.
I'm so happy to be here withyou today.
Speaker 1 (03:09):
Great to be here.
Joel, Happy to talk supplychain with you in the next 45
minutes.
Speaker 2 (03:17):
Always good to talk about supply chain.
You obviously have a veryimpressive background.
I'm not sure I told you this inadvance.
I am a failed academic, if youwill.
I started the path of anacademic career in economics,
ended up dropping out, movingfull time into technology.
You've had a very impressivecareer.
(03:38):
I know MIT is one of theleading institutions around
supply chain.
I'm curious to start out howdid you end up in this space?
Speaker 1 (03:47):
Actually, this is a great question, joel.
I started 30 years ago atColumbia University as an
assistant professor, focusing ondoing a lot of theoretical work
around the analysis ofanalytics for vehicle routing
problems.
Most of the work was purelytheoretical but at some point in
(04:11):
the first couple of years Ireceived a call from the New
York City Board of Education.
This is think about the timing.
This is about 1993.
The person on the other side ofthe call is telling me we are
developing a new decisionsupport system for school bus
(04:32):
routing.
We would like to work with youon the analytics, on the engine
that will generate the routesthat the school buses will
follow.
I was not that interested.
I was doing theory.
I was focusing on my academiccareer.
I expressed this concern to theperson on the other side.
(04:57):
At some point there was a pauseand the person was asking me
are you interested in making abig impact?
Do you want to be relevant?
I'm thinking here is myopportunity to transfer the
theories that I developed intopractical algorithm.
(05:19):
I started collaborating with theBoard of Education in New York
in the development of a newsystem.
Remember the time existed in1993.
As a technology that we seetoday was not available at the
time.
We needed to develop theanalytics.
Then we needed to connect theanalytics, the route generated
(05:40):
by the analytics, with digitalmaps show, present the routes
generated to the planners, makesure that the routes follow all
the requirements, the safetyrequirements, introduced by the
Board of Education.
All of a sudden we developed asystem that was used in the
(06:03):
entire New York City anddramatic improvement in
performance of the system.
This technology won the bestwindow-developed technology
award that was given byMicrosoft in 1994, 1995.
All of a sudden I realized, hey, I can develop a system that
(06:28):
uses analytics as the mainengine but requires way beyond
the analytics.
But I was looking where are theopportunities to make an impact
?
How many cities you have thatare the size of New York?
Very few, and so I did not knowanything about supply chain.
And all of a sudden I realizedthere is this new emerging area
(06:52):
called supply chain where maybethis analytics can make an
impact.
And I basically founded thefirst company called logic tools
that focused on supply chainanalytics.
And this company that wasfounded in 1996, 1997 had three
types of technologies one forsupply chain network design, one
(07:17):
for multi-ethic loan inventoryoptimization and the third one
for production planning andscheduling.
Ten years later, think about thetiming.
We had 350 clients using ourtechnology on a day-to-day basis
.
We saw the company and becamepart of IBM technology
infrastructure in 2008, and thenI was.
(07:41):
I joined IBM, helping them withthe new technology, and then in
2011, I founded a businessanalytics company.
Think about the timing.
This is a time where nobodytalked about application of AI
and machine learning forbusinesses.
We founded one of the firstcompanies that focus on business
(08:04):
analytics for operation and andand supply chain, and very
quickly I followed this up witha cloud technology company that
provided opportunities fordevelopers and for companies to
implement to customize theirsolution to their own
(08:24):
environment.
These two companies thebusiness analytics and the cloud
technology companies became apart of Accenture technology
infrastructure.
I'm company free.
I don't push any company, butright now I lead the MIT data
science lab, and let me tell youjust a little bit about the MIT
(08:47):
data science lab.
That that I lead the MIT datascience lab is a partnership
between MIT and about 25companies, focusing on some of
the most challenging problemsthat these companies have by
bringing to together data modelsand analysis.
(09:08):
We have companies, partnersfrom different industries,
high-tech, cpg, finance, evengovernment, are part of the
partnership, as well as softwarecompanies, and we have a global
footprint and we covervarieties of areas.
One of the areas that we coverand I will end up with this
(09:32):
story is the area of supplychain resiliency.
Everybody, as you know verywell, joel today is excited and
focused on supply chainresiliency.
We at the lab started focusingon supply chain resiliency way
before the pandemic in 2011,2012, after events like the
tsunami in Japan and the floodin Thailand, and, if you
(09:55):
remember, the volcano eruptionin Iceland in 2010, and we were
fortunate enough to collaboratewith the Ford Motor Company,
develop a new way to measure theresiliency of any supply chain,
identify hidden risk anddevelop mitigation strategies.
(10:19):
Our technology, the MIT datascience lab technology, was
implemented by Ford a cosy intersupply chain.
We received the bestengineering technology
implemented at Ford in 2015 andwe published a few technical
papers, but the ones that I willrefer to here is the Harvard
(10:42):
Business Review article that waspublished in 2014.
Between the publication of thisarticle in 2014 focused on
executive and supply chainleaders.
In the beginning of thepandemic, Joel, I was knocking
on many doors trying to convinceexecutives and leaders to focus
(11:04):
on supply chain resiliency verylittle interest.
Most companies, as you knowvery well, focus on supply chain
efficiencies, on cutting costsin the supply chain.
Everything changed at thebeginning of the pandemic, but
perhaps not in the way youexpect.
Think January 2020.
(11:24):
The pandemic was in China, itwas not in North America, it was
not in Europe.
I started collecting data onwhat's happening in China and I
realized I can use thetechnology I developed six years
earlier to understand theimpact of what's happening in
China on supply chain in NorthAmerica and in Europe, and I
very quickly published a paperin February of 2020 in Harvard
(11:50):
Business Review saying supplychain in the title and in the
first paragraph supply chain inNorth America and Europe will
stop by mid-March.
And this is precisely this isexactly what happened.
March 17, fortune magazinereported the entire automotive
industry in Europe is shuttingdown.
(12:10):
The New York time, march 182020, reported the entire
automotive industry in the US,in Mexico, in Canada, is
shutting down, and after that wegot an enormous interest in our
technology.
Different consulting companiesstarted implementing what we
develop.
(12:31):
Others have used the conceptthat we developed, emphasizing
that they focus on supply chainresiliency, and I will end up
just with a short comment whichsays last year, in May of 2022,
the US president economic reporthad a section on our technology
(12:56):
, recommending to companies andencouraging companies to use
supply chain stress test thatwhat I called we call our
technology to make sure they areready for the next disruption.
We all you myself, yourcolleague, we are all supply
chain professional.
Think about this.
Speaker 2 (13:15):
The highest decision-maker in the country
recognize that what we all arefocusing on is important for the
national security of thecountry well, they do say timing
is everything and you'veobviously been studying and
focusing on supply chain for,you know, decades at this, this
(13:35):
point, and you're veryinteresting to walk through that
timeline.
You, as you say, you knowpublishing those, seeing this,
this wave of disruption comingand then having it happen.
Before we jump into thatarticle, because I definitely
want to talk about that, I'dencourage anyone listening to to
read it, a fantastic articlethat you put in Harvard Business
(13:55):
Review.
How fast did you see thatchange happen when COVID hit?
I mean, was it just all of asudden, you know, your inbox was
flooded and phone calls werecoming in, or did it take people
a while to you know realize themagnitude, even when COVID hit,
of what was going to happen?
Speaker 1 (14:10):
it was relatively quiet until the publication of
our paper in middle of February.
And then the first call that Igot was before mid-March, a call
from Ford.
The people I collaborated withsix years earlier said hey, we
read your article.
Nobody understood why you arepredicting mid-March.
(14:33):
Can you come and give a talk?
And I think March 2nd, on March3rd I don't remember the exact
day I gave a talk at Ford withhundreds of people participating
from the entire organizationtrying to understand.
Hey, these guys that workedwith us Seekers earlier, he's
saying he used the same concept,the same technology to predict.
(14:55):
Two weeks later I get anotherset of calls from, called from
Ford people hey, the predictionwas highly accurate.
Speaker 2 (15:03):
And after that was a flood of calls and interest in
what we did so let's talk alittle bit about, or maybe a lot
about, supply chaindigitization, because that's
really kind of the overarchingtheme of your article.
There's many different elementshere that we'll dive into, and
I know that you've been broughtinto a lot of companies you know
(15:23):
to talk about this and whatFolks can you just start?
I'm sure people here arefamiliar with the term.
A lot of consulting firms areout there pushing supply chain
digitization.
Can you explain what it is andwhy it's important?
It's always been important, butit's probably particularly
important today.
Speaker 1 (15:42):
Again, great observation and a very good
question to start the discussionaround supply chain
digitization, george, and maybeas a background, we need to talk
about perception versus realitywhen we focus on supply chain
digitization.
If you think about theperception before the pandemic,
(16:03):
the perception that executivehad about supply chain
digitization was that itrequires significant financial
investment.
It will take many years peoplewere talking about five to six
years to digitize the entiresupply chain because it requires
(16:24):
instrumentation of everyprocess and every product and
every facilities.
What we learned during thepandemic is that the future is
here, that with moderatefinancial investment, by taking
advantage of existing data thatcompanies have and complimenting
(16:47):
it with additional data thatcompanies can get from third
party partners, we can achievealmost all the benefits of
full-flight supply chaindigitization in a relatively
short period of time.
Remember, the perception wasthis is five, six years.
Now we are talking about 12 to18 months.
(17:09):
The perception is that we needenormous financial investment.
Now we know we don't.
Why, because the key aboutsupply chain digitization is to
focus on a few key capabilities.
What are the key capabilities?
The first is what I highlightin the Harvard Business Review
(17:31):
article, which is all aboutunified view of demand.
Taking traditional consensusforecast that most companies are
using, what is the consensusforecast?
In consensus forecast,different functional areas will
generate their own forecast.
Finance has a forecast, saleshas a forecast, supply chain has
(17:52):
a forecast.
A trade, the people who areresponsible for marketing,
pricing and discounting hastheir own forecast.
And then they come together ina consensus meeting to agree on
a compromise.
In supply chain digitization,we replace this process which is
mostly manual process and alsothe compromise does not
(18:14):
necessarily match with marketdemand with the process where we
agree on the data and then letthe analytic generate a forecast
that is used by its differentfunctional areas, a different
level of granularity.
So that's one importantcapability.
The second important capabilityin supply chain digitization is
(18:35):
to replace one size fits allsupply chain strategy with a
segmented supply chain strategy.
What is a segmented supplychain strategy?
We recognize that differentproduct has different
characteristics, differentchannels have different
objectives and differentlimitations and, as a result,
what we want to do is to segment, to cluster product channels,
(19:01):
market into segments that insome sense are similar and for
each segment we generate its ownsupply chain strategy.
And finally, on top of this, webuild synergies across the
different segments in order toleverage economies of scale and
to reduce complexity.
So that's the second capability.
(19:22):
First is unified view of demandreplacing consensus forecast.
Second, replacing one size fitsall with a segmented supply
chain strategy.
Once we have these twocapabilities, we can take
advantage of them in smart SNOP.
Snop is a process thatcompanies have used you know
(19:42):
very well, everybody knows verywell have been using since the
late 1980s to continuouslybalance supply and demand.
Now we take advantage of theunified view of demand and SNOP
is really one integrated processwith segmentation and with
(20:03):
unified view of demand.
And the last element toemphasize and I will stop after
that is as good as our plan is,there are always deviation from
the plan supply disruption,changes in demand.
How can I quickly identifythose deviations and disruption
(20:24):
and respond effectively to thosechallenges?
This is what I refer to in thearticle as smart execution.
What cuts it causes fordifferent capabilities is
digitization, analytics andautomation, and to do that, you
want to take advantage ofunified or a single platform
(20:49):
that allow you to aggregate,allow you to normalize and allow
you to visualize the data.
Speaker 2 (20:55):
If I've understood you here, it's like those first
three points that unified viewof demand, the kind of having
segmentation one size doesn'tfit all and then that smart
planning.
Those all flow into let'screate the best possible plan
that we can.
And then the last point is nomatter how well we plan, there's
(21:19):
always going to be deviations.
That plan is wrong the momentit's printed and so we ought to
be able to adjust, and Icertainly agree with that.
We see that a lot, and I wantto drill into some of these a
little bit.
Let's talk about this unifiedview of demand, and you spoke
about the way companies tend toplan today, this kind of
(21:39):
consensus-driven process, eachteam bringing something forward.
Do you see what you'reproposing?
Is this about speeding up thatprocess?
Is it about removing human biasfrom the process?
Maybe it's both, maybe it'sneither.
(22:01):
Maybe there are other elements.
I mean, where do you see thebiggest opportunities and the
biggest gains here?
Speaker 1 (22:07):
So if we dive into unified view of demand, there
are elements from each of thecomments that you made and maybe
it will be an effective way toillustrate unified view of
demand and its impact byconsidering one recent
(22:28):
implementation at a largeconsumer-packaged good
manufacturing company.
This is a large company that weknow well, that you know very
well.
It's a global company servingmany, many retailers, thousands
of SKUs, many distributioncenters and manufacturing
(22:51):
facilities.
The characteristic of thesupply chain before the
transformation was very typicalto CPG.
It was all one size fits allconsensus forecast, manual
processes, and the company wasvery good at squeezing costs
(23:16):
from different processes,focusing on improving the
efficiency of differentprocesses in the supply chain.
But when you look into it, yourealize inventory is high,
service level is not necessarilywhere they want it to be, waste
is high as well, and so thequestion was how can we change
that?
And the main focus of unifiedview of demand was to change the
(23:44):
consensus forecast that isinternally focused into a
consumption-driven supply chainstrategy.
Let's repeat this is startingwith consumption and generating
a focus of what the retailerwill order from the manufacturer
(24:05):
.
Now how do you achieve thatobjective?
First, you focus on fourdifferent data sources.
The first data element is allabout the internal data that the
company has Orders that theyreceive from retailers, product
(24:27):
characteristics and so forth andso on.
The second is consumptioninformation about demand faced
by retailers, and here typicallyyou want to have point-of-sell
data.
The problem that this companyhas is that most retailers
refuse to give thempoint-of-sell data, and so here
(24:50):
we use the third-party data fromcompanies like IRI and Nielsen,
companies that we are allfamiliar with.
The third was all aboutmicro-economics data inflation,
unemployment because this helpedus understand customer demand.
And the last element was allsorts of Google trends, social
(25:14):
media information, socialnetwork information that allow
us to better understand demand,especially for a new product
introduction.
It's these four sources of datathat allow us to start by
predicting what the retailerwill face in the market.
Think about this this is theCPG generate its own prediction
(25:39):
of what retailer will face inthe market, retailer by retailer
, skew by skew, week by week inthe case that we are focusing on
for the next 80 weeks, and thisprediction was an input into an
engine that generate a forecastof what the retailer will order
(26:01):
from the CPG.
So the forecast of what theretailer will order from the CPG
is driven by two importantelements the historical orders
from the retailer to the CPGthat's obvious, that everybody's
using, together with forecastof what the retailer will face
in the market, skew by skew,retailer by retailer, week by
(26:26):
week, for the next 80 weeks.
And all of a sudden a processthat was very manual and involve
human making adjustment to theforecast became much more
automated, where the focus ofhuman is to understand the data
and make sure that we agree onthe data that we are using.
(26:49):
And one element that I'm notsure always follows, but one
element that was part of theprocess was don't sell though
anybody else don't come and say,hey, I don't like the forecast
increasing by 5% or decreasingby 10%.
We had an incident when salescame and said look, the forecast
(27:09):
in the West Coast is way higherthan our intuition and
experience suggests.
That's all good, let's look atthe data and see why we are
wrong.
And in that case it turns outthat the number of stores that
was documented in the data setwhere the product is being sold
in the West Coast was way higherthan the numbers that sales
(27:36):
people knew were product areselling.
Adjusting for the numbersreduce the gap between their
intuition and what came out ofthe tool.
The point is it's all theelements that you mentioned.
It's automated, but it involvespeople in the process, human in
the process to make sure thatwe use the right data to
(27:59):
generate a focus.
Speaker 2 (28:01):
I love that story that you just shared because
I've certainly seen this in mycareer.
I imagine you have a lot whenyou you get into analytics, data
science, whatever you want tocall it, I guess these days you
bring the data and you run into,I guess, that intuition, as you
call it.
You know people who say I'vebeen selling this product for 30
years.
I have a pretty good gutreaction and it can't.
(28:23):
There can be a lot of conflictthere where people you know
they'll try to Identify problemswith the data and so let's just
throw that all out, let's relyon my intuition.
And and I loved your example ofthat hey, instead of saying
let's bump that number fivepercent or let's change it,
really trying to root cause whatis?
You know?
(28:44):
People have a lot of ideas.
That intuition comes fromsomewhere and sometimes people
know things that aren'treflected in those models.
But let's bring that intuitioninto the models, rather than
kind of throwing it out and justsaying let's just arbitrarily,
you know, make changes herewithout describing why they are.
Speaker 1 (29:01):
So I think that Example of the number of
locations a really good one andand maybe to support your point,
one thing that I typicallyhighlight in discussion around
data science, analytics, unifiedview of demand, supply chain
digitization, is that zero tosuccess is not about just about
(29:24):
science.
When we see a killeropportunity, huge impact is when
we bring together science andArt.
Science is all the analyticsand the data and the algorithms
that we are talking about.
Art is the intuition, theexperience, the domain knowledge
(29:45):
that business as supply chainprofessional have.
Where I see biggest impact iswhen we combine the two into An
important opportunity.
The second thing, just tocontinue along the lines that
you highlighted, is this is notjust about science and art.
This is not just aboutdeveloping Effective focus.
(30:08):
This is also and I'm sure youhave seen this before this is
also about being able to explainwhy is the focus, is the way it
is.
Nobody put in a different wayand no decision-makers that I'm
familiar with is going to lookat the forecast coming out of a
blackboard, say, hey, this isgreat, let's use it tomorrow.
(30:32):
They want to understand what'shappening.
And when you think aboutexplainability, there are
multiple levels ofexplainability that you need to
address, and part of thecapability today is the ability
to provide that insight.
The explainability has threedimension.
Let's enumerate them tohighlight what is important and
(30:55):
what can be done.
The first people want tounderstand what drives my
forecast.
They look at your forecast, theforecast you generated or I
generated, and say, hey, I seethat this skew or this product
family is going to do very wellin the Midwest In the summer.
Explain to why?
Is it?
Because our pricing, ourmarketing or just the product
(31:17):
itself.
So Identifying, explaining thedriver of the forecast is one
part of the story, but this isnot the only one.
The second one and the secondone I'm sure again, you and your
client have seen this before Isa comment coming from finance.
You generate a forecast, youget a call from finance and and
(31:40):
the father, the person on theother side of the call, is
telling you hey, you gave me aforecast for December.
That is, forecast for Decemberis different than the forecast
for December you gave me lastmonth.
What change?
Why do I see 5% in December?
It should have been exactly thesame as the forecast you gave
me a month ago.
We need to make sure that weunderstand what drove the chain
(32:03):
in the forecast that we are,because if if we cannot explain
it, nobody will use it.
I Right.
And the last element is to beable to explain the difference
between the forecast and reality, between the forecast and what
we observe in the market.
These are, to me, the threeimportant capabilities that any
(32:26):
effective Demand forecast toolneeds to have to make sure that
people trust and use theforecast.
Speaker 2 (32:35):
So let's change gears a little bit and let's talk
about what you mentioned at thebeginning around supply chain
resiliency.
You know a topic that you know.
I think you mentioned this inyour article.
You know these competinginterests between that kind of
efficiency, I think for the lastcouple decades.
You know, with globalization,every company how do I, how do I
squeeze down costs and you knowthat can compete to some degree
(32:59):
with Resiliency, making surethat I can weather.
You know, anything that comesup.
There have been supply chaindisruptions forever, but they
probably haven't been in thenews like they have the last
couple years.
So curious to get your yourthoughts here on the right
balance between those two andhow you know companies what,
what steps they should be takingto make sure that they have
(33:21):
more resiliency in their supplychain.
Speaker 1 (33:24):
So.
So let's start with what arethe key capabilities in supply
chain resiliency that weemphasize, that we found to be
important when Companies isfocusing not only on efficiency
but also on resiliency, andthere are three capabilities.
The first one is planning.
(33:45):
I want to be able to model mysupply chain, analyze the supply
chain and identify hidden riskand develop integration
strategies Right.
That's one important capabilitythat now is recognized by
companies as an important, as anarea of Focus, and I've seen
(34:05):
this surprisingly CPG high-techautomotive it's now a cost the
board Because of what happenedover the last couple of years.
The second one is Is monitoring.
Nothing is wrong, but I want tounderstand how much exposure to
risk I have in my supply chainso that something goes wrong, I
(34:29):
will know where my potentialproblems are.
Think about this much likefinancial Investment, companies
are measuring their exposure ona day-to-day basis for two,
financial risk.
Why exposure to supply chainrisk is changing from week to
week.
Lead time is increasing,inventory depleted.
(34:49):
I want to understand how andwhere there are changes in my
exposure and, if it's belowcertain threshold, maybe
expedite or maybe change myinventory strategy.
And the last element isresponding.
Something happened.
There is an earthquake inThailand, one of my supplier
(35:11):
facilities is affected.
Right now I have enough formaterial to feed my assembly
line.
However, because of thesedisruptions that happened
yesterday, six weeks from now Imay not have enough for material
to feed the assembly line.
If I know this today, I will beable to respond effectively.
And if you think about the storyI told you about the beginning
(35:34):
of the pandemic, that's actuallywhat I did.
I Basically use the technologyto understand hey, something is
happening in China, when is itgoing to hit the supply chain in
North America and Europe?
And if you did the same thingat the beginning of the pandemic
, you would have secured rawmaterial, a component for your
supply chain, to make sure thatyou can feed the supply chain on
(35:57):
an ongoing basis.
All of these capabilities theplanning, the monitoring and the
responding rely on Data andAnalytics.
But one important element thatwe found in implementation of
this across multiple companieseven before the pandemic supply
(36:19):
chain resiliency, surprisingly,is not necessarily going against
Efficiency or cost cutting.
So in the implementation atFord, it's all in the paper, so
there is nothing that that isconfidential in the
implementation at Ford, we foundthat for knowing much through a
(36:42):
lot of inventory on thecomponent Because we had a way
to measure the resiliency oftheir supply chain.
We identify cost savingsopportunity.
We were able to reduce or Fordwas able to reduce inventory by
certain percentage point andstill maintain a resilient
(37:03):
Supply chain.
So supply chain resiliency isdefinitely about identifying
hidden risk and developingmitigation strategies, but it's
also about Identifying wherethey built too much inventory
and where are the opportunitiesto cut inventory While still
maintaining a resilient supplychain.
Speaker 2 (37:25):
Yeah, I appreciate you sharing that, because You're
right, and even the way I framethe question is that those
things are at odds.
You get efficiency or you getresiliency, and so that's very
interesting.
You know that example at Fordand presumably at many other
companies, there would beopportunities to really do both.
So fascinating.
I also I liked your analogy oftalking about finance.
(37:49):
I previously was in financialtechnology, so I have a little
background there and I do thinkof all these firms that have
risk management functions andthey're obviously constantly
monitoring the portfolio, whatthe risk is.
Doing that much more real timeand if you're running a supply
chain, you should be doing thesame thing.
But I want to go back tosomething you mentioned at the
beginning around when we startedtalking about supply chain
(38:11):
digitization, and you mentionedthat it can be intimidating if
you think of this as being afive, six year project, millions
and millions of dollars thatcan be overwhelming, and that's
something we hear a lot, youknow.
Just, it's like how do I getstarted?
And you mentioned that you knowgetting started might be a lot
easier, faster and cheaper thancompanies expected, and so
(38:36):
curious to get your thoughts onhow companies do start down that
path.
Speaker 1 (38:41):
That's again a very good way to discuss my
experience, at least incompanies that focus on supply
chain digitization, and theexperience that I have is that
you don't start with full fledgeimplementation the way I just
(39:02):
described here.
The focus that I've seen so fara lot of time is all about a
proof of concept.
That takes six to seven weeksthat allow you to demonstrate
the value, illustrate some ofthe capabilities and maybe
(39:23):
convince yourself and yourcolleague that you should go in
a digital transformation journey.
And let me highlight what Ithink the key element in such a
proof of concept.
So a proof of concept is allabout assessing where you are
(39:45):
Remember I told you the storyabout the CPG company, some of
the challenges that they had.
So assessing where they are,identifying where you want to be
what people in industry callthe north star and develop a
plan for closing the gap.
In this plan, you want todemonstrate value.
(40:08):
You want to illustrate with asmall data set what is the
improvement you can achieve onfocus accuracy, and we never do
focus accuracy improvement forthe sake of the focus.
We do it to reduce inventory,to increase service level, to
increase revenue, to cut costs.
If you can demonstrate thatthen you have an opportunity to
(40:33):
open the door for a digitalsupply chain transformation in
this process.
You also want to develop theoperating model, suggest how are
we going to execute?
Execute the digitized strategy.
How do we make sure that theprocess sticks?
(40:53):
Remember unified view of theend.
You want to make sure that atthe end of this process, the
different functional area do notgo back to their own department
saying, ah, the old guy iswrong.
We will continue using thefocus that we generated using
consensus model, using the oldstatistical model.
Speaker 2 (41:17):
So identifying process and identify
organizational structure thatmake sure that this new strategy
is executed effectively is partof the six, seven weeks initial
exercise now, okay, thank you,and I like your comment about
(41:38):
because we hear this a lotpeople saying I want to be more
accurate with my forecasting andit's like, well, yeah, we all
do, but for what reason?
What are you going to do withthat?
How's that going to, you know,reduce inventory costs, or, you
know, increase sales, orwhatever it may be.
So I think, always aligning itto where the dollars, so as we
close, can you tell, tell me andand our listeners, a little
(42:02):
more about the data science labat MIT the data science lab, as
I mentioned, is a partnershipwith 25 companies.
Speaker 1 (42:13):
The areas that we cover start with supply chain
resiliency we talked about itearlier supply chain
digitization but we do waybeyond that.
We do a lot of work in thespace of price optimization.
We did work of priceoptimization for one of the
largest online fashion retailerin Europe, a company called the
(42:38):
lando, where every week, ouranalytics what comes out of the
data science lab price 1.5million different skills in 23
different countries.
We do it for online platformand we do it for brick and
mortar retailers a company likecouple, which is a large
retailer in Mexico, companies inthe Middle East.
(43:04):
We do a lot of work inpersonalized offering and let me
explain this with an example.
If you fly today, joel, fromLondon to Paris, when you
purchase your airline ticket, weare not involved, but once you
purchase the airline ticket, oursystem kicks in and offer you
(43:24):
an salary, product, priorityboarding, current hotel, and the
offers that you will get, joel,may be different than the
offers that I will get, becausethe system is learning about
individual preferences.
We have done a lot of work oninventory, transportation and
procurement, especially lastmile delivery for a software
(43:46):
company like Blue Yonder.
So some of our partner aresoftware companies that
collaborate with us in a waythat we develop the engine and
they implemented as part oftheir tools excellent, and if
someone wanted to get in touchwith the data science lab or
with you, what would be the bestway to do?
that just send me an email,dslavymitedu and I can provide
(44:12):
more information about what wedo, about our partners and about
the impact of our technology ona variety of companies
excellent.
Speaker 2 (44:23):
Well, david, I so appreciate you joining us today
and all your insights.
Thank you, you've beenlistening to professor David
Simchi-Levy, professor ofengineering systems at MIT and
head of the MIT data science lab.
That's all for this week.
See you next time on ChefLife.
Before the pandemic, the perception that executive
had about supply chaindigitization was that it
required significant financialinvestment.
It will take many years peoplewere talking about five to six
years to digitize the entiresupply chain, because it
(00:23):
required instrumentation ofevery process and every product
and every facilities.
What we learned during thepandemic is that the future is
here, that with moderatefinancial investment, by taking
advantage of existing data thatcompanies have and complementing
(00:47):
it with additional data thatcompanies can get from
third-party partners, we canachieve almost all the benefits
of full-flight supply chaindigitization in a relatively
short period of time.
Speaker 2 (01:04):
From LAI.
This is Shelf Life.
What is supply chaindigitization and why does it
(01:25):
matter now?
How can you better use the datayou already have to increase
your forecast accuracy?
How can you reduce supply chainrisks in a time of constant
disruptions?
On every episode of Shelf Life,we answer questions like these
and more, with the help ofleaders across the consumer
goods industry.
Today, I'm very happy towelcome Professor David Simche
(01:47):
Levy, Professor of EngineeringSystems at MIT and head of the
MIT Data Science Lab.
David is widely considered oneof the leading voices in supply
chain management, havingpublished four books on the
subject and countless articlesDuring his career.
He's also founded severalcompanies focused on analytics
and supply chain optimization.
I'm your host, Joel Beale, CEOof Aloe AI.
(02:10):
We'll be back with David rightafter this.
Speaker 3 (02:29):
There's plenty of data to help, but pulling and
analyzing reports from retailerportals is so tedious and time
consuming that you can't respondto problems at the shelf until
the bull whip hits you in theface.
That's cow, actually.
Now there's Aloeai to help.
Aloe automatically aggregatesand harmonizes data from your
retailers, supply chain partnersand ERP.
(02:50):
Then we make it easy to findinsights using pre-built
consumer goods, specific metricsand dashboards, so you can
sense, predict and respondinstantly.
Check it out today and get ademo at Aloeai.
Speaker 2 (03:04):
David, welcome to Shelf Life.
I'm so happy to be here withyou today.
Speaker 1 (03:09):
Great to be here.
Joel, Happy to talk supplychain with you in the next 45
minutes.
Speaker 2 (03:17):
Always good to talk about supply chain.
You obviously have a veryimpressive background.
I'm not sure I told you this inadvance.
I am a failed academic, if youwill.
I started the path of anacademic career in economics,
ended up dropping out, movingfull time into technology.
You've had a very impressivecareer.
(03:38):
I know MIT is one of theleading institutions around
supply chain.
I'm curious to start out howdid you end up in this space?
Speaker 1 (03:47):
Actually, this is a great question, joel.
I started 30 years ago atColumbia University as an
assistant professor, focusing ondoing a lot of theoretical work
around the analysis ofanalytics for vehicle routing
problems.
Most of the work was purelytheoretical but at some point in
(04:11):
the first couple of years Ireceived a call from the New
York City Board of Education.
This is think about the timing.
This is about 1993.
The person on the other side ofthe call is telling me we are
developing a new decisionsupport system for school bus
(04:32):
routing.
We would like to work with youon the analytics, on the engine
that will generate the routesthat the school buses will
follow.
I was not that interested.
I was doing theory.
I was focusing on my academiccareer.
I expressed this concern to theperson on the other side.
(04:57):
At some point there was a pauseand the person was asking me
are you interested in making abig impact?
Do you want to be relevant?
I'm thinking here is myopportunity to transfer the
theories that I developed intopractical algorithm.
(05:19):
I started collaborating with theBoard of Education in New York
in the development of a newsystem.
Remember the time existed in1993.
As a technology that we seetoday was not available at the
time.
We needed to develop theanalytics.
Then we needed to connect theanalytics, the route generated
(05:40):
by the analytics, with digitalmaps show, present the routes
generated to the planners, makesure that the routes follow all
the requirements, the safetyrequirements, introduced by the
Board of Education.
All of a sudden we developed asystem that was used in the
(06:03):
entire New York City anddramatic improvement in
performance of the system.
This technology won the bestwindow-developed technology
award that was given byMicrosoft in 1994, 1995.
All of a sudden I realized, hey, I can develop a system that
(06:28):
uses analytics as the mainengine but requires way beyond
the analytics.
But I was looking where are theopportunities to make an impact
?
How many cities you have thatare the size of New York?
Very few, and so I did not knowanything about supply chain.
And all of a sudden I realizedthere is this new emerging area
(06:52):
called supply chain where maybethis analytics can make an
impact.
And I basically founded thefirst company called logic tools
that focused on supply chainanalytics.
And this company that wasfounded in 1996, 1997 had three
types of technologies one forsupply chain network design, one
(07:17):
for multi-ethic loan inventoryoptimization and the third one
for production planning andscheduling.
Ten years later, think about thetiming.
We had 350 clients using ourtechnology on a day-to-day basis
.
We saw the company and becamepart of IBM technology
infrastructure in 2008, and thenI was.
(07:41):
I joined IBM, helping them withthe new technology, and then in
2011, I founded a businessanalytics company.
Think about the timing.
This is a time where nobodytalked about application of AI
and machine learning forbusinesses.
We founded one of the firstcompanies that focus on business
(08:04):
analytics for operation and andand supply chain, and very
quickly I followed this up witha cloud technology company that
provided opportunities fordevelopers and for companies to
implement to customize theirsolution to their own
(08:24):
environment.
These two companies thebusiness analytics and the cloud
technology companies became apart of Accenture technology
infrastructure.
I'm company free.
I don't push any company, butright now I lead the MIT data
science lab, and let me tell youjust a little bit about the MIT
(08:47):
data science lab.
That that I lead the MIT datascience lab is a partnership
between MIT and about 25companies, focusing on some of
the most challenging problemsthat these companies have by
bringing to together data modelsand analysis.
(09:08):
We have companies, partnersfrom different industries,
high-tech, cpg, finance, evengovernment, are part of the
partnership, as well as softwarecompanies, and we have a global
footprint and we covervarieties of areas.
One of the areas that we coverand I will end up with this
(09:32):
story is the area of supplychain resiliency.
Everybody, as you know verywell, joel today is excited and
focused on supply chainresiliency.
We at the lab started focusingon supply chain resiliency way
before the pandemic in 2011,2012, after events like the
tsunami in Japan and the floodin Thailand, and, if you
(09:55):
remember, the volcano eruptionin Iceland in 2010, and we were
fortunate enough to collaboratewith the Ford Motor Company,
develop a new way to measure theresiliency of any supply chain,
identify hidden risk anddevelop mitigation strategies.
(10:19):
Our technology, the MIT datascience lab technology, was
implemented by Ford a cosy intersupply chain.
We received the bestengineering technology
implemented at Ford in 2015 andwe published a few technical
papers, but the ones that I willrefer to here is the Harvard
(10:42):
Business Review article that waspublished in 2014.
Between the publication of thisarticle in 2014 focused on
executive and supply chainleaders.
In the beginning of thepandemic, Joel, I was knocking
on many doors trying to convinceexecutives and leaders to focus
(11:04):
on supply chain resiliency verylittle interest.
Most companies, as you knowvery well, focus on supply chain
efficiencies, on cutting costsin the supply chain.
Everything changed at thebeginning of the pandemic, but
perhaps not in the way youexpect.
Think January 2020.
(11:24):
The pandemic was in China, itwas not in North America, it was
not in Europe.
I started collecting data onwhat's happening in China and I
realized I can use thetechnology I developed six years
earlier to understand theimpact of what's happening in
China on supply chain in NorthAmerica and in Europe, and I
very quickly published a paperin February of 2020 in Harvard
(11:50):
Business Review saying supplychain in the title and in the
first paragraph supply chain inNorth America and Europe will
stop by mid-March.
And this is precisely this isexactly what happened.
March 17, fortune magazinereported the entire automotive
industry in Europe is shuttingdown.
(12:10):
The New York time, march 182020, reported the entire
automotive industry in the US,in Mexico, in Canada, is
shutting down, and after that wegot an enormous interest in our
technology.
Different consulting companiesstarted implementing what we
develop.
(12:31):
Others have used the conceptthat we developed, emphasizing
that they focus on supply chainresiliency, and I will end up
just with a short comment whichsays last year, in May of 2022,
the US president economic reporthad a section on our technology
(12:56):
, recommending to companies andencouraging companies to use
supply chain stress test thatwhat I called we call our
technology to make sure they areready for the next disruption.
We all you myself, yourcolleague, we are all supply
chain professional.
Think about this.
Speaker 2 (13:15):
The highest decision-maker in the country
recognize that what we all arefocusing on is important for the
national security of thecountry well, they do say timing
is everything and you'veobviously been studying and
focusing on supply chain for,you know, decades at this, this
(13:35):
point, and you're veryinteresting to walk through that
timeline.
You, as you say, you knowpublishing those, seeing this,
this wave of disruption comingand then having it happen.
Before we jump into thatarticle, because I definitely
want to talk about that, I'dencourage anyone listening to to
read it, a fantastic articlethat you put in Harvard Business
(13:55):
Review.
How fast did you see thatchange happen when COVID hit?
I mean, was it just all of asudden, you know, your inbox was
flooded and phone calls werecoming in, or did it take people
a while to you know realize themagnitude, even when COVID hit,
of what was going to happen?
Speaker 1 (14:10):
it was relatively quiet until the publication of
our paper in middle of February.
And then the first call that Igot was before mid-March, a call
from Ford.
The people I collaborated withsix years earlier said hey, we
read your article.
Nobody understood why you arepredicting mid-March.
(14:33):
Can you come and give a talk?
And I think March 2nd, on March3rd I don't remember the exact
day I gave a talk at Ford withhundreds of people participating
from the entire organizationtrying to understand.
Hey, these guys that workedwith us Seekers earlier, he's
saying he used the same concept,the same technology to predict.
(14:55):
Two weeks later I get anotherset of calls from, called from
Ford people hey, the predictionwas highly accurate.
Speaker 2 (15:03):
And after that was a flood of calls and interest in
what we did so let's talk alittle bit about, or maybe a lot
about, supply chaindigitization, because that's
really kind of the overarchingtheme of your article.
There's many different elementshere that we'll dive into, and
I know that you've been broughtinto a lot of companies you know
(15:23):
to talk about this and whatFolks can you just start?
I'm sure people here arefamiliar with the term.
A lot of consulting firms areout there pushing supply chain
digitization.
Can you explain what it is andwhy it's important?
It's always been important, butit's probably particularly
important today.
Speaker 1 (15:42):
Again, great observation and a very good
question to start the discussionaround supply chain
digitization, george, and maybeas a background, we need to talk
about perception versus realitywhen we focus on supply chain
digitization.
If you think about theperception before the pandemic,
(16:03):
the perception that executivehad about supply chain
digitization was that itrequires significant financial
investment.
It will take many years peoplewere talking about five to six
years to digitize the entiresupply chain because it requires
(16:24):
instrumentation of everyprocess and every product and
every facilities.
What we learned during thepandemic is that the future is
here, that with moderatefinancial investment, by taking
advantage of existing data thatcompanies have and complimenting
(16:47):
it with additional data thatcompanies can get from third
party partners, we can achievealmost all the benefits of
full-flight supply chaindigitization in a relatively
short period of time.
Remember, the perception wasthis is five, six years.
Now we are talking about 12 to18 months.
(17:09):
The perception is that we needenormous financial investment.
Now we know we don't.
Why, because the key aboutsupply chain digitization is to
focus on a few key capabilities.
What are the key capabilities?
The first is what I highlightin the Harvard Business Review
(17:31):
article, which is all aboutunified view of demand.
Taking traditional consensusforecast that most companies are
using, what is the consensusforecast?
In consensus forecast,different functional areas will
generate their own forecast.
Finance has a forecast, saleshas a forecast, supply chain has
(17:52):
a forecast.
A trade, the people who areresponsible for marketing,
pricing and discounting hastheir own forecast.
And then they come together ina consensus meeting to agree on
a compromise.
In supply chain digitization,we replace this process which is
mostly manual process and alsothe compromise does not
(18:14):
necessarily match with marketdemand with the process where we
agree on the data and then letthe analytic generate a forecast
that is used by its differentfunctional areas, a different
level of granularity.
So that's one importantcapability.
The second important capabilityin supply chain digitization is
(18:35):
to replace one size fits allsupply chain strategy with a
segmented supply chain strategy.
What is a segmented supplychain strategy?
We recognize that differentproduct has different
characteristics, differentchannels have different
objectives and differentlimitations and, as a result,
what we want to do is to segment, to cluster product channels,
(19:01):
market into segments that insome sense are similar and for
each segment we generate its ownsupply chain strategy.
And finally, on top of this, webuild synergies across the
different segments in order toleverage economies of scale and
to reduce complexity.
So that's the second capability.
(19:22):
First is unified view of demandreplacing consensus forecast.
Second, replacing one size fitsall with a segmented supply
chain strategy.
Once we have these twocapabilities, we can take
advantage of them in smart SNOP.
Snop is a process thatcompanies have used you know
(19:42):
very well, everybody knows verywell have been using since the
late 1980s to continuouslybalance supply and demand.
Now we take advantage of theunified view of demand and SNOP
is really one integrated processwith segmentation and with
(20:03):
unified view of demand.
And the last element toemphasize and I will stop after
that is as good as our plan is,there are always deviation from
the plan supply disruption,changes in demand.
How can I quickly identifythose deviations and disruption
(20:24):
and respond effectively to thosechallenges?
This is what I refer to in thearticle as smart execution.
What cuts it causes fordifferent capabilities is
digitization, analytics andautomation, and to do that, you
want to take advantage ofunified or a single platform
(20:49):
that allow you to aggregate,allow you to normalize and allow
you to visualize the data.
Speaker 2 (20:55):
If I've understood you here, it's like those first
three points that unified viewof demand, the kind of having
segmentation one size doesn'tfit all and then that smart
planning.
Those all flow into let'screate the best possible plan
that we can.
And then the last point is nomatter how well we plan, there's
(21:19):
always going to be deviations.
That plan is wrong the momentit's printed and so we ought to
be able to adjust, and Icertainly agree with that.
We see that a lot, and I wantto drill into some of these a
little bit.
Let's talk about this unifiedview of demand, and you spoke
about the way companies tend toplan today, this kind of
(21:39):
consensus-driven process, eachteam bringing something forward.
Do you see what you'reproposing?
Is this about speeding up thatprocess?
Is it about removing human biasfrom the process?
Maybe it's both, maybe it'sneither.
(22:01):
Maybe there are other elements.
I mean, where do you see thebiggest opportunities and the
biggest gains here?
Speaker 1 (22:07):
So if we dive into unified view of demand, there
are elements from each of thecomments that you made and maybe
it will be an effective way toillustrate unified view of
demand and its impact byconsidering one recent
(22:28):
implementation at a largeconsumer-packaged good
manufacturing company.
This is a large company that weknow well, that you know very
well.
It's a global company servingmany, many retailers, thousands
of SKUs, many distributioncenters and manufacturing
(22:51):
facilities.
The characteristic of thesupply chain before the
transformation was very typicalto CPG.
It was all one size fits allconsensus forecast, manual
processes, and the company wasvery good at squeezing costs
(23:16):
from different processes,focusing on improving the
efficiency of differentprocesses in the supply chain.
But when you look into it, yourealize inventory is high,
service level is not necessarilywhere they want it to be, waste
is high as well, and so thequestion was how can we change
that?
And the main focus of unifiedview of demand was to change the
(23:44):
consensus forecast that isinternally focused into a
consumption-driven supply chainstrategy.
Let's repeat this is startingwith consumption and generating
a focus of what the retailerwill order from the manufacturer
(24:05):
.
Now how do you achieve thatobjective?
First, you focus on fourdifferent data sources.
The first data element is allabout the internal data that the
company has Orders that theyreceive from retailers, product
(24:27):
characteristics and so forth andso on.
The second is consumptioninformation about demand faced
by retailers, and here typicallyyou want to have point-of-sell
data.
The problem that this companyhas is that most retailers
refuse to give thempoint-of-sell data, and so here
(24:50):
we use the third-party data fromcompanies like IRI and Nielsen,
companies that we are allfamiliar with.
The third was all aboutmicro-economics data inflation,
unemployment because this helpedus understand customer demand.
And the last element was allsorts of Google trends, social
(25:14):
media information, socialnetwork information that allow
us to better understand demand,especially for a new product
introduction.
It's these four sources of datathat allow us to start by
predicting what the retailerwill face in the market.
Think about this this is theCPG generate its own prediction
(25:39):
of what retailer will face inthe market, retailer by retailer
, skew by skew, week by week inthe case that we are focusing on
for the next 80 weeks, and thisprediction was an input into an
engine that generate a forecastof what the retailer will order
(26:01):
from the CPG.
So the forecast of what theretailer will order from the CPG
is driven by two importantelements the historical orders
from the retailer to the CPGthat's obvious, that everybody's
using, together with forecastof what the retailer will face
in the market, skew by skew,retailer by retailer, week by
(26:26):
week, for the next 80 weeks.
And all of a sudden a processthat was very manual and involve
human making adjustment to theforecast became much more
automated, where the focus ofhuman is to understand the data
and make sure that we agree onthe data that we are using.
(26:49):
And one element that I'm notsure always follows, but one
element that was part of theprocess was don't sell though
anybody else don't come and say,hey, I don't like the forecast
increasing by 5% or decreasingby 10%.
We had an incident when salescame and said look, the forecast
(27:09):
in the West Coast is way higherthan our intuition and
experience suggests.
That's all good, let's look atthe data and see why we are
wrong.
And in that case it turns outthat the number of stores that
was documented in the data setwhere the product is being sold
in the West Coast was way higherthan the numbers that sales
(27:36):
people knew were product areselling.
Adjusting for the numbersreduce the gap between their
intuition and what came out ofthe tool.
The point is it's all theelements that you mentioned.
It's automated, but it involvespeople in the process, human in
the process to make sure thatwe use the right data to
(27:59):
generate a focus.
Speaker 2 (28:01):
I love that story that you just shared because
I've certainly seen this in mycareer.
I imagine you have a lot whenyou you get into analytics, data
science, whatever you want tocall it, I guess these days you
bring the data and you run into,I guess, that intuition, as you
call it.
You know people who say I'vebeen selling this product for 30
years.
I have a pretty good gutreaction and it can't.
(28:23):
There can be a lot of conflictthere where people you know
they'll try to Identify problemswith the data and so let's just
throw that all out, let's relyon my intuition.
And and I loved your example ofthat hey, instead of saying
let's bump that number fivepercent or let's change it,
really trying to root cause whatis?
You know?
(28:44):
People have a lot of ideas.
That intuition comes fromsomewhere and sometimes people
know things that aren'treflected in those models.
But let's bring that intuitioninto the models, rather than
kind of throwing it out and justsaying let's just arbitrarily,
you know, make changes herewithout describing why they are.
Speaker 1 (29:01):
So I think that Example of the number of
locations a really good one andand maybe to support your point,
one thing that I typicallyhighlight in discussion around
data science, analytics, unifiedview of demand, supply chain
digitization, is that zero tosuccess is not about just about
(29:24):
science.
When we see a killeropportunity, huge impact is when
we bring together science andArt.
Science is all the analyticsand the data and the algorithms
that we are talking about.
Art is the intuition, theexperience, the domain knowledge
(29:45):
that business as supply chainprofessional have.
Where I see biggest impact iswhen we combine the two into An
important opportunity.
The second thing, just tocontinue along the lines that
you highlighted, is this is notjust about science and art.
This is not just aboutdeveloping Effective focus.
(30:08):
This is also and I'm sure youhave seen this before this is
also about being able to explainwhy is the focus, is the way it
is.
Nobody put in a different wayand no decision-makers that I'm
familiar with is going to lookat the forecast coming out of a
blackboard, say, hey, this isgreat, let's use it tomorrow.
(30:32):
They want to understand what'shappening.
And when you think aboutexplainability, there are
multiple levels ofexplainability that you need to
address, and part of thecapability today is the ability
to provide that insight.
The explainability has threedimension.
Let's enumerate them tohighlight what is important and
(30:55):
what can be done.
The first people want tounderstand what drives my
forecast.
They look at your forecast, theforecast you generated or I
generated, and say, hey, I seethat this skew or this product
family is going to do very wellin the Midwest In the summer.
Explain to why?
Is it?
Because our pricing, ourmarketing or just the product
(31:17):
itself.
So Identifying, explaining thedriver of the forecast is one
part of the story, but this isnot the only one.
The second one and the secondone I'm sure again, you and your
client have seen this before Isa comment coming from finance.
You generate a forecast, youget a call from finance and and
(31:40):
the father, the person on theother side of the call, is
telling you hey, you gave me aforecast for December.
That is, forecast for Decemberis different than the forecast
for December you gave me lastmonth.
What change?
Why do I see 5% in December?
It should have been exactly thesame as the forecast you gave
me a month ago.
We need to make sure that weunderstand what drove the chain
(32:03):
in the forecast that we are,because if if we cannot explain
it, nobody will use it.
I Right.
And the last element is to beable to explain the difference
between the forecast and reality, between the forecast and what
we observe in the market.
These are, to me, the threeimportant capabilities that any
(32:26):
effective Demand forecast toolneeds to have to make sure that
people trust and use theforecast.
Speaker 2 (32:35):
So let's change gears a little bit and let's talk
about what you mentioned at thebeginning around supply chain
resiliency.
You know a topic that you know.
I think you mentioned this inyour article.
You know these competinginterests between that kind of
efficiency, I think for the lastcouple decades.
You know, with globalization,every company how do I, how do I
squeeze down costs and you knowthat can compete to some degree
(32:59):
with Resiliency, making surethat I can weather.
You know, anything that comesup.
There have been supply chaindisruptions forever, but they
probably haven't been in thenews like they have the last
couple years.
So curious to get your yourthoughts here on the right
balance between those two andhow you know companies what,
what steps they should be takingto make sure that they have
(33:21):
more resiliency in their supplychain.
Speaker 1 (33:24):
So.
So let's start with what arethe key capabilities in supply
chain resiliency that weemphasize, that we found to be
important when Companies isfocusing not only on efficiency
but also on resiliency, andthere are three capabilities.
The first one is planning.
(33:45):
I want to be able to model mysupply chain, analyze the supply
chain and identify hidden riskand develop integration
strategies Right.
That's one important capabilitythat now is recognized by
companies as an important, as anarea of Focus, and I've seen
(34:05):
this surprisingly CPG high-techautomotive it's now a cost the
board Because of what happenedover the last couple of years.
The second one is Is monitoring.
Nothing is wrong, but I want tounderstand how much exposure to
risk I have in my supply chainso that something goes wrong, I
(34:29):
will know where my potentialproblems are.
Think about this much likefinancial Investment, companies
are measuring their exposure ona day-to-day basis for two,
financial risk.
Why exposure to supply chainrisk is changing from week to
week.
Lead time is increasing,inventory depleted.
(34:49):
I want to understand how andwhere there are changes in my
exposure and, if it's belowcertain threshold, maybe
expedite or maybe change myinventory strategy.
And the last element isresponding.
Something happened.
There is an earthquake inThailand, one of my supplier
(35:11):
facilities is affected.
Right now I have enough formaterial to feed my assembly
line.
However, because of thesedisruptions that happened
yesterday, six weeks from now Imay not have enough for material
to feed the assembly line.
If I know this today, I will beable to respond effectively.
And if you think about the storyI told you about the beginning
(35:34):
of the pandemic, that's actuallywhat I did.
I Basically use the technologyto understand hey, something is
happening in China, when is itgoing to hit the supply chain in
North America and Europe?
And if you did the same thingat the beginning of the pandemic
, you would have secured rawmaterial, a component for your
supply chain, to make sure thatyou can feed the supply chain on
(35:57):
an ongoing basis.
All of these capabilities theplanning, the monitoring and the
responding rely on Data andAnalytics.
But one important element thatwe found in implementation of
this across multiple companieseven before the pandemic supply
(36:19):
chain resiliency, surprisingly,is not necessarily going against
Efficiency or cost cutting.
So in the implementation atFord, it's all in the paper, so
there is nothing that that isconfidential in the
implementation at Ford, we foundthat for knowing much through a
(36:42):
lot of inventory on thecomponent Because we had a way
to measure the resiliency oftheir supply chain.
We identify cost savingsopportunity.
We were able to reduce or Fordwas able to reduce inventory by
certain percentage point andstill maintain a resilient
(37:03):
Supply chain.
So supply chain resiliency isdefinitely about identifying
hidden risk and developingmitigation strategies, but it's
also about Identifying wherethey built too much inventory
and where are the opportunitiesto cut inventory While still
maintaining a resilient supplychain.
Speaker 2 (37:25):
Yeah, I appreciate you sharing that, because You're
right, and even the way I framethe question is that those
things are at odds.
You get efficiency or you getresiliency, and so that's very
interesting.
You know that example at Fordand presumably at many other
companies, there would beopportunities to really do both.
So fascinating.
I also I liked your analogy oftalking about finance.
(37:49):
I previously was in financialtechnology, so I have a little
background there and I do thinkof all these firms that have
risk management functions andthey're obviously constantly
monitoring the portfolio, whatthe risk is.
Doing that much more real timeand if you're running a supply
chain, you should be doing thesame thing.
But I want to go back tosomething you mentioned at the
beginning around when we startedtalking about supply chain
(38:11):
digitization, and you mentionedthat it can be intimidating if
you think of this as being afive, six year project, millions
and millions of dollars thatcan be overwhelming, and that's
something we hear a lot, youknow.
Just, it's like how do I getstarted?
And you mentioned that you knowgetting started might be a lot
easier, faster and cheaper thancompanies expected, and so
(38:36):
curious to get your thoughts onhow companies do start down that
path.
Speaker 1 (38:41):
That's again a very good way to discuss my
experience, at least incompanies that focus on supply
chain digitization, and theexperience that I have is that
you don't start with full fledgeimplementation the way I just
(39:02):
described here.
The focus that I've seen so fara lot of time is all about a
proof of concept.
That takes six to seven weeksthat allow you to demonstrate
the value, illustrate some ofthe capabilities and maybe
(39:23):
convince yourself and yourcolleague that you should go in
a digital transformation journey.
And let me highlight what Ithink the key element in such a
proof of concept.
So a proof of concept is allabout assessing where you are
(39:45):
Remember I told you the storyabout the CPG company, some of
the challenges that they had.
So assessing where they are,identifying where you want to be
what people in industry callthe north star and develop a
plan for closing the gap.
In this plan, you want todemonstrate value.
(40:08):
You want to illustrate with asmall data set what is the
improvement you can achieve onfocus accuracy, and we never do
focus accuracy improvement forthe sake of the focus.
We do it to reduce inventory,to increase service level, to
increase revenue, to cut costs.
If you can demonstrate thatthen you have an opportunity to
(40:33):
open the door for a digitalsupply chain transformation in
this process.
You also want to develop theoperating model, suggest how are
we going to execute?
Execute the digitized strategy.
How do we make sure that theprocess sticks?
(40:53):
Remember unified view of theend.
You want to make sure that atthe end of this process, the
different functional area do notgo back to their own department
saying, ah, the old guy iswrong.
We will continue using thefocus that we generated using
consensus model, using the oldstatistical model.
Speaker 2 (41:17):
So identifying process and identify
organizational structure thatmake sure that this new strategy
is executed effectively is partof the six, seven weeks initial
exercise now, okay, thank you,and I like your comment about
(41:38):
because we hear this a lotpeople saying I want to be more
accurate with my forecasting andit's like, well, yeah, we all
do, but for what reason?
What are you going to do withthat?
How's that going to, you know,reduce inventory costs, or, you
know, increase sales, orwhatever it may be.
So I think, always aligning itto where the dollars, so as we
close, can you tell, tell me andand our listeners, a little
(42:02):
more about the data science labat MIT the data science lab, as
I mentioned, is a partnershipwith 25 companies.
Speaker 1 (42:13):
The areas that we cover start with supply chain
resiliency we talked about itearlier supply chain
digitization but we do waybeyond that.
We do a lot of work in thespace of price optimization.
We did work of priceoptimization for one of the
largest online fashion retailerin Europe, a company called the
(42:38):
lando, where every week, ouranalytics what comes out of the
data science lab price 1.5million different skills in 23
different countries.
We do it for online platformand we do it for brick and
mortar retailers a company likecouple, which is a large
retailer in Mexico, companies inthe Middle East.
(43:04):
We do a lot of work inpersonalized offering and let me
explain this with an example.
If you fly today, joel, fromLondon to Paris, when you
purchase your airline ticket, weare not involved, but once you
purchase the airline ticket, oursystem kicks in and offer you
(43:24):
an salary, product, priorityboarding, current hotel, and the
offers that you will get, joel,may be different than the
offers that I will get, becausethe system is learning about
individual preferences.
We have done a lot of work oninventory, transportation and
procurement, especially lastmile delivery for a software
(43:46):
company like Blue Yonder.
So some of our partner aresoftware companies that
collaborate with us in a waythat we develop the engine and
they implemented as part oftheir tools excellent, and if
someone wanted to get in touchwith the data science lab or
with you, what would be the bestway to do?
that just send me an email,dslavymitedu and I can provide
(44:12):
more information about what wedo, about our partners and about
the impact of our technology ona variety of companies
excellent.
Speaker 2 (44:23):
Well, david, I so appreciate you joining us today
and all your insights.
Thank you, you've beenlistening to professor David
Simchi-Levy, professor ofengineering systems at MIT and
head of the MIT data science lab.
That's all for this week.
See you next time on ChefLife.
Popular Podcasts
Stuff You Should Know
If you've ever wanted to know about champagne, satanism, the Stonewall Uprising, chaos theory, LSD, El Nino, true crime and Rosa Parks, then look no further. Josh and Chuck have you covered.
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations. Follow now to get the latest episodes of Dateline NBC completely free, or subscribe to Dateline Premium for ad-free listening and exclusive bonus content: DatelinePremium.com
Las Culturistas with Matt Rogers and Bowen Yang
Ding dong! Join your culture consultants, Matt Rogers and Bowen Yang, on an unforgettable journey into the beating heart of CULTURE. Alongside sizzling special guests, they GET INTO the hottest pop-culture moments of the day and the formative cultural experiences that turned them into Culturistas. Produced by the Big Money Players Network and iHeartRadio.