Jack Levis An interview with - Putting Analytics in the Driver's Seat
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
An interview with
Jack Levis
Director of Process Management at UPS
Putting Analytics in the Driver’s Seat
Transform to the power of digitalJack Levis
goal is to turn that complex universe of Capgemini Consulting: You spoke
data into business intelligence. of prescriptive analytics playing an
effective role in route optimization.
Let me give you an example. We have Can you give us more details on
about 55,000 package car drivers in the its role in overcoming your key
challenges?
US alone and around 106,000 drivers,
globally, for our entire vehicle fleet, Jack Levis: We have implemented
and we deliver more than 16 million a number of prescriptive analytics
packages daily. When you consider the projects across our business but the
fact that every driver at UPS has trillions one that stands out from the rest is our
of ways to run their delivery routes, route optimization program, based on
the number of possibilities increases prescriptive analytics, called ORION
exponentially. However, not all of these (On-Road Integrated Optimization and
Jack Levis routes are necessarily optimal in terms
of fuel efficiency and distance. Consider
Navigation). We formally started the
ORION project in 2003 and began
Director of Process Management the fact that a reduction of one mile per to roll out the system in 2012. We are
at UPS driver per day translates to savings of very serious about using prescriptive
up to $50 million a year. The question modeling for our routes. So much so,
becomes: how do you mine the sea of that we have 500 people dedicated
data from our sensors and vehicles to to ORION. In fact, ORION is probably
A reduction of one arrive at the most effective route for our one of the largest prescriptive analytics
drivers? systems ever deployed.
mile per driver per day
translates to savings of Digitizing Operations
up to $50 million a year.
Capgemini Consulting: How did
Our analytics system
UPS start its digital transformation enabled UPS to
Capgemini Consulting: Can you journey?
start by giving us a background
Jack Levis: Our digital journey
eliminate 85 million
to UPS and some of the unique
challenges that a logistics player
started with an early adoption of data miles driven per year.
of your size faces? and analytics tools for improving our
operations. As our operations became
Jack Levis: UPS is a business that more complex and distributed in nature,
thrives on managing complexity. the focus has been to improve business As I mentioned earlier, effectively mining
processes, increase efficiency and the sea of data from our sensors and
Meeting our high levels of customer cut costs. We had been following a vehicles to arrive at the most effective
service entails complexity. We not only descriptive1 and predictive analytics2- route for our drivers is a huge challenge.
aim to deliver every package on time, based system for a long time but what It is here that our prescriptive analytics
but we provide customers with multiple has recently changed is our shift to system shines. It hides this large
service options to meet their needs. We prescriptive analytics3. I can safely say amount of alternative routes while giving
even allow adjusting of delivery choices that UPS is one of the few companies to drivers clear inputs, thereby taking the
while the shipment is in route. Executing effectively use prescriptive analytics to guesswork out of the equation. The
this mission means constantly gain insight for successful optimization. best part is that the system produces
orchestrating orders, adjusting route these answers in as little as six to eight
schedules and following up on package seconds. The idea is not to make big
deliveries with a massive fleet of ground changes in driver routes. In fact, the
and air vehicles. This exercise generates Our digital journey optimized route might look very similar to
huge amounts of data feeds, from the driver’s normal route. However, the
devices, vehicles, tracking materials
started with an early real benefit lies in the distance it helps
and sensors. Each of these feeds also adoption of analytics reduce – a quarter mile shaved here and
comes with its own data format. Our a half mile shaved there. So, the system
tools. keeps looking for ways to deliver minute
savings throughout the day.Jack Levis
How Analytics Transformed
Operations at UPS
The Logistical Business Benefits
Complexities at UPS of Analytics
55,000 Reduction of
package car drivers 85 million
in US alone miles driven/ year
106,000
drivers globally
8 million fewer
gallons of fuel used
16 million
packages daily
Reduction in engine idling
One driver = trillions of ways time by 10 million
to run delivery routes minutes
Shaving just one mile/ driver Reduction in carbon C02
= $50 million footprint by 6,500
savings a year metric tons
The challenge: Arriving at the Analytics has changed the way
most optimal route for drivers UPS functionsJack Levis
Capgemini Consulting: What evaluated. We then look at the best In the past, we used metrics that
have been the tangible benefits way to meet those needs, and often showed incremental change from year
that you realized by deploying analytics is needed. to year. We looked at things such as
analytics systems in your number of deliveries made per hour, or
operations? Our goal is to make business processes,
the amount of time expended for a route
methods, procedures, and analytics all
Jack Levis: The deployment of vs. a work measurement standard.
one in the same. For the front line user,
descriptive and predictive analytics Those used to be the measure of
the use of analytics results becomes
systems several years ago enabled success for deliveries. But now, after
just part of the job.
UPS to reduce 85 million miles driven implementing our prescriptive analytics
per year. That equates to over 8 million With systems that require large process system, the metrics have become far
fewer gallons of fuel used. Prescriptive change, we spend significant effort more sophisticated and nuanced. We
analytics adds to those gains. ensuring that the change can actually have moved from looking at lagging
be attained. This often requires iterative indicators that focus on end results only
We were also able to reduce engine
prototyping so that we can successfully to looking at leading indicators.
idling time by 10 million minutes. This led
achieve the business gains.
to significant savings in fuel consumption We created balanced scorecards that
– around 650,000 gallons – and we have We followed this approach for our guide the front-line operators on areas
reduced our carbon emissions by over prescriptive analytics system, ORION. to focus. The elements have been
6,500 metric tons. We tested the system over two years. carefully selected and weighted. The
We had to prove that the program balanced scorecard measure correlates
Now adding to this, deploying efficient
would, indeed, measurably impact highly with true business results.
prescriptive analytics systems has
costs. Not only were the numbers were
enabled UPS to eliminate miles from our
impressive, but the fact that front line
routes in 2013. The surprising fact is that
operators and drivers were supportive
we have realized this additional benefit
with only 18% of UPS delivery routes
got everyone’s attention. That helped The key to managing
convince our senior management to test
deployed.
the program at other locations across change in roll out of
the country. After testing the system in
15 different locations, final approval for
digital initiatives is
broad-based deployment of the initiative to take a collaborative
across the company was given.
As we deployed approach.
analytics, we realized
Capgemini Consulting: As part
we could not continue of implementing analytics across
your operations, did you have to Capgemini Consulting: What
relying on old metrics. change the way you traditionally were some of the biggest issues
looked at your metrics? UPS faced when implementing its
Jack Levis: An important thing to note prescriptive analytics program?
about analytics systems – especially Jack Levis: As I mentioned earlier, our
Implementing and prescriptive analytics – is that change analytics system has been built to tell us
management is required. New ways
Measuring Digital of operating are being produced and
the best delivery route for a particular
day. If you consider that there are going
front line employees must be educated to be 55,000 different drivers, which
Capgemini Consulting: What and supported. This means changing means 55,000 different routes the
is your approach to launching behavior. model has to work for – that is a very
analytics initiatives? hard model to build. We would have our
As we deployed analytics, we realized
Jack Levis: The business drives that we often could not continue to parameters and all the dials tuned and
technology at UPS. We don’t look at measure a new way of doing business the answer would be great on Tuesday.
initiatives as ‘analytics projects’, we look with the same old metrics. So, we had to But the same system with Wednesday’s
at them as business projects. Before come up with new metrics that enabled data would not work. So, definitely the
launching an initiative, areas where effective measurement. model needed to have a lot of heuristics,
the greatest business need exist are math, and business rules built in. That
was a big challenge.Jack Levis
We spent years making our data better It is because of this collaborative
and changing the algorithm so it wasn’t approach that our drivers are included
so sensitive to changes in data. We had in the process, which is why there Our challenge hasn’t
to have an algorithm and process that hasn’t been much resistance to change
didn’t take a rocket scientist to use. In a when implementing analytics initiatives.
been around identifying
step-by-step matter we tested not only In many cases, drivers have said “my analytics talent as
that the algorithm could be created, but stress is reduced”. This is because
that we could transfer the knowledge to the system makes thousands of small much as it has been
the front line. decisions for them, freeing up the driver
to make the larger decisions of servicing
in determining the
Operational the customer. best way to train the
Excellence
hundreds of business
Capgemini Consulting: With the
Capgemini Consulting: significant shortage of digital skills people who are using
Implementation of digital initiatives across industries, how does UPS
entails large-scale change acquire its analytics talent? these tools.
management. How did you
convince drivers who relied on Jack Levis: Our challenge hasn’t been
traditional route planning to shift to around identifying analytics talent as
new analytical tools? much as it has been in determining
the best way to train the hundreds of
Jack Levis: We adopt a highly
collaborative approach with our drivers
business people who are using these Digital Future
tools. For this, we provide role-based
in implementing these initiatives. We
training that teaches employees how
are acutely aware that our drivers aren’t Capgemini Consulting: Looking
to use the analytics system. When fully ahead, what are some of the new
automatons who rely on insights from
deployed, the system will offer our front- digital initiatives UPS is working
the analytics system. When you have
line supervisors and drivers the tools to on to further drive operational
a prescriptive analytics system, usually
test scenarios and make tradeoffs. They efficiency?
problems arise because of issues with
don’t need to be data experts, but they
data. Counteracting bad data requires Jack Levis: Our current analytics
need to understand which parameters
collaboration with our drivers. This is systems are still largely static in nature
impact which performance objective.
why we have built-in buffers into our – we need to change that. They do not
Drivers are graphically shown how the
model where we acknowledge that account for unexpected situations, such
algorithm is deriving different parts of
the system is not perfect and that as traffic delays or accidents. In such
the route so they can compare it with
drivers have the opportunity to identify scenarios, our drivers are expected to
their own experience and attempt to
flaws. We let our drivers exercise their take a discretionary call. In future, we
beat it.
discretion. We tell them: if the model has anticipate moving from a static to a
you doing something that won’t meet a In our analytics team, we also have dynamic manifest. For this, we are trying
customer’s demand, do what’s right. people with business backgrounds to make our plans more flexible and with
who understand the overall system provisions for real-time updates.
objectives from an organizational
perspective. So, while an analytics
person will sift through algorithms, the
We now want to use our software engineer will translate it into
prescriptive analytics code, the business person will ensure
that the solution meets the desired
system to offer objectives.
innovative services to
our customers.Jack Levis
Capgemini Consulting: How do you As we go in to the future, we hope to
perceive analytics adding further spin-off many new services based on
value to route efficiency at UPS? real-time updates in the system. We Our goal is to move
Jack Levis: Our current focus has want to be able to offer customers to
been in trying to reduce the distance make last-minute requests. Calculating from our current static
the costs of last-minute request changes
covered during each delivery. We will
from customers is another aspect that
analytics systems to a
continue to do that. But we also want
to use our prescriptive analytics system an analytics system can deliver. This dynamic manifest.
to offer more innovative services to our would enable us to possibly reschedule
customers. We have already opened or re-prioritize deliveries based on
up our internal supply chain to our several scenarios.
customers to enable them to make
specific pick-up/drop requests.Jack Levis
References
1 Descriptive analytics refers to a set of techniques used to describe or explore or profile any kind of data.
2 Predictive analytics encompasses a variety of techniques that analyze current and historical facts to make
predictions about future, or otherwise unknown, events.
3 Prescriptive analytics represents the final phase of business analytics, which mines data to suggest decision options to take
advantage of a future opportunity or mitigate a future risk.Jack Levis
Jack Levis
Director of Process Management at UPS
UPS is a global package delivery company headquartered in Atlanta, USA. The company
operates in over 220 countries with over 399,000 employees. In 2012, it had over 8.8
million customers with delivery volume of some 4.1 billion. The company generated $54
billion in revenue in 2012. UPS has been at the forefront of deploying advanced analytics in
optimizing its operations. Capgemini Consulting spoke with Jack Levis, Director of Process
Management at UPS.
About Capgemini
Capgemini Consulting is the global strategy and transformation With around 120,000 people in 40 countries, Capgemini is one
consulting organization of the Capgemini Group, specializing of the world’s foremost providers of consulting, technology
in advising and supporting enterprises in significant and outsourcing services. The Group reported 2011 global
transformation, from innovative strategy to execution and with revenues of EUR 9.7 billion. Together with its clients, Capgemini
an unstinting focus on results. With the new digital economy creates and delivers business and technology solutions that
creating significant disruptions and opportunities, our global fit their needs and drive the results they want. A deeply
team of over 3,600 talented individuals work with leading multicultural organization, Capgemini has developed its own
companies and governments to master Digital Transformation, way of working, the Collaborative Business ExperienceTM, and
drawing on our understanding of the digital economy and draws on Rightshore®, its worldwide delivery model.
our leadership in business transformation and organizational
change. Learn more about us
at www.capgemini.com.
Find out more at:
http://www.capgemini-consulting.com/
Rightshore® is a trademark belonging to Capgemini
Contacts Didier Bonnet, didier.bonnet@capgemini.com Jerome Buvat, jerome.buvat@capgemini.com
Capgemini Consulting is the strategy and transformation consulting brand of Capgemini Group. The information contained in this document is proprietary.
© 2013 Capgemini. All rights reserved.You can also read
