Cloud, saas and analytics driven value chain business transformation version 1.1
1. Salil Amonkar
Value Chain Business Transformation
Industry Expert and Thought Leader
The convergence of Cloud, SaaS and
Machine Learning represents a new
opportunity to drive significant Business
Transformation in the Value Chain
Cloud, SaaS and Machine
Learning
for
Value Chain
Transformation
2. Contents
2 Title of the book
3
4
5
8
10
11
12
Note from the author
Fundamentals of Value Chain Best Practices
Impact of Cloud SaaS and Analytics on Value Chain
Characteristics of solutions for Value Chain Business
Transformation
Review of Market Place Solutions
Summary
References
3. Note from the author
3 Cloud, SaaS and Advanced Analytics for Value Chain Business
Transformation
Dear Readers,
It is my privilege to share with you insights from the business transformational experience
I have gained working over nearly three decades with various High-Tech companies in the
Silicon Valley, Discrete Manufacturing and Retail companies in Europe and Asia.
The pace of change brought forth by the disruptive effect of cloud and SaaS is occurring at
a much faster pace than that can be matched by enterprises. The high-tech industry has gone
through multiple transformations, from mainframe to PC to client server architecture during the
pre Y2K days, to the post-Y2K era of the Internet and now, with Cloud and SaaS. With this
evolution now data can be securely accessible anytime, anywhere by anyone. Expectations
around when, how and how much the customer should expect to pay for all of this has also
seen tremendous change.
Example of this change can be noted in how high tech companies are forced to react in
terms of their offerings and the monetization models that they need to focus on. An example of
this change in the value chain context is retail companies trying to rapidly build their
Omnichannel capabilities as they realize that driving a customer to a decision in their favor
rests on a brand’s ability to engage the customer on multiple fronts all with a consistent
customer experience. These and other dramatic shifts on the business fronts will drive most
enterprises to look at opportunities to ensure that their Value Chain processes are able to keep
up with this change.
In my role as Management Consultant focusing on leveraging Technology to drive Business
Transformation and having worked with several companies in implementing successful
business transformational initiatives I have gained a few insights that I would like to share with
all. I have used these insights in driving our own internal Cloud SaaS product design which has
been primarily based on the needs expressed by our customers that we see as remaining
unfulfilled.
My organization Pluto7 focuses on business transformation for Value Chain through Cloud
SaaS and Analytics and we have developed our own cloud-based supply chain management
subscription offering - Planning in a Box. Our collective experience has also contributed to
thoughts expressed in this paper.
- Salil Amonkar
4. Fundamentals of Value Chain Best Practices
The fundamentals of Value Chain Planning and Supply Chain Planning can be applied conceptually to
all industries. These are the four key areas namely Plan, Source, Make and Deliver, except that for
service supply chains the make and deliver is replaced by serve and deliver.
Many publications have been written by many thought leaders and experts on Top 10 Best Practices
for Value Chain and Supply Chain Management. After having read many of these and, coupled with
my experience of having successfully deployed these as solutions across multiple customers, the
following represents my personal view on how these apply to the above five areas:
1. An effective Value Chain is one where the corporation passes value on to the customer, with the
least possible cost to the business.
2. Plan is the starting point and driver of the Value Chain/Supply Chain Planning processes and
how this is managed has cascading impact on rest of the Value Chain. Best practices around
Plan process area have to do with Forecasting, Collaboration and Metrics.
3. Best practice(s) around Sourcing Operations are key to having the necessary flexibility to follow
demand changes while meeting lead times and cost constraints.
4. Make implementations across the industry have focused on best practices like Lean and Agile
Manufacturing and usage of Technology to drive automation in factory management and related
logistics.
5. Finally but not the least the effective external collaboration between partners, delivery
scheduling and streamlined logistics (forward and reverse logistics) are key to a good delivery
process. Cloud SaaS and Analytical solutions using Predictive Analytics and
Machine Learning provide a disruptive and significant potential
when used innovatively to drive Forecasting with high forecast
accuracy, easy collaboration and provide business real time data for
KPI management.
Similarly using Data Sciences and Machine Learning to determine
supply patterns can be used to achieve flexible sourcing operations
at lower cost.
Finally Machine Learning and Artificial Intelligence provide
capabilities to streamline Distribution and Logistics processes in a
way that has not been possible before.
Fundamentals of Value
Chain Best Practices
conceptually do not
change, what changes is
how technology advances
can be leveraged to
achieve the same
4 Cloud, SaaS and Advanced Analytics for Value Chain Business Transformation
5. Figuratively speaking Value Chain (Supply Chain) is the
infrastructure like combination of roads, train tracks, air
routes, sea routes , stations, airports etc. and Forecasting sets
the plan on how organizations will move traffic through these.
However as in case of real highway experiences that we go
through in everyday life; similarly weaknesses in forecasting
cause slow downs and impeded movements in the Supply
Chain.
The key to Forecasting is learning from past trends. At the
heart of machine learning is an algorithm that "learns" from
data which can be used to drive ever more accurate
predictors of demand which in turn will drive value chain
planning. This game-changing capability is possibly the
opportunity to reduce costs and reduce confusion. Using
Supply Chain jargon it can significantly help reduce the
impact of the “Bull Whip” effect which causes small demand
changes to have significant impact on back end supply chain.
Using artificial intelligence derived from machine
learning based models to establish a variation over the
normal baseline that is typically generated by today’s
forecasting models enable exception based demand
management leading to higher forecast accuracy. This is done
by ensuring that appropriate visualizations are provided that
clearly highlight the exceptions.
Using Cloud SaaS based approach to facilitate feeding of
multiple forecast input data feeds in a collaborative manner
helps enhance the forecast accuracy since more data and
more variations enable the machine learning models to be
even more effective.
5
Plan – Forecasting, Collaboration
& KPIs
Machine learning based Artificial Intelligence
models applied to forecast data and coupled
with appropriate visualizations that highlight
only those demand items that need to be
managed help drive forecast accuracy. It
provides the ability to react to quick changes
in demand that otherwise cannot be detected
by today’s conventional methods. Using KPI’s
such as Forecast Accuracy, Forecast Bias is
recommended.
Cloud, SaaS and Advanced Analytics for Value Chain Business
Transformation
Impact of Cloud SaaS and Analytics on Value Chain
Forecasting is
learning from
past trends.
Machine learning
with algorithms
that learn from
data can be used
to drive accurate
prediction of
Forecast which in
turn can drive
Value Chain
6. Impact of Cloud SaaS and Analytics on Value Chain
6 Cloud, SaaS and Advanced Analytics for Value Chain Business
Transformation
The Technology advances in Cloud SaaS and Advanced
Analytics based on Artificial Intelligence models based on
machine learning have significant impact on transforming the
Source, Make and Deliver Functions in the Value Chain.
While having a good Sourcing Strategy is an important
piece in the Supply Chain its operational implementation is
dependent upon significantly on usage of technology.
Effective collaboration is key for sourcing and effectiveness of
current on premise sourcing solutions is largely dependent
upon the integration of such solutions across enterprises.
Typically these have come at a cost resulting in not all
partners being able to leverage their functionality and thus
causing gaps in collaboration. Typical example of these are
gaps in consigned inventory, visibility of customer owned
inventory and so on.
Cloud SaaS solutions enable cost effective approach to
bridge these gaps. In addition the ability to use the large data
collected in the sourcing value chain and process it with
machine learning models enables the detection of trends that
can then be compared with demand management leading to
improved ability to react to changes and avoid costly
situations. This is done by not only leveraging the Cloud SaaS
Advanced Analytics solution to determine the trends but also
enable efficient and timely collaboration within all partners in
the Sourcing Value Chain through actionable information
available anywhere (i.e. desktop, mobile).
Cloud SaaS Advanced
Analytical solutions
not only improve
ability to react to
changes in demand
but also enable
efficient
collaboration within
Sourcing Partners
with actionable
information available
on desktop and
mobile
7. Impact of Cloud SaaS and Analytics on Value Chain
7 Cloud, SaaS and Advanced Analytics for Value Chain Business
Transformation
The Technology advances in Cloud SaaS and Advanced Analytics based on Artificial
Intelligence models combined with the developments in the internet of things and internet of
everything space is making the fully automated factory of the future a possible reality today.
IOE-IOT sensors sensing inputs from inventory in multiple locations such as in-transit,
consigned sites, supplier docks, receiving stores, supply carousels, distribution centers feeding
into the manufacturing execution systems, operations planning systems running machine
learning driven artificial intelligence algorithms can help manage optimized inventory levels.
Similarly taking sensor data from assembly lines and using machine learning to not only
achieve finely tuned statistical process control (quality control) but also predict potential
failures before they even occur thus eliminating potential for costly shut downs, rework and
obtain significant productivity on the shop floor is now possible.
Similar to sourcing which focuses more on inbound logistics Cloud SaaS solutions enable
cost effective approach to bridge the collaboration barrier between Manufacturing and
Distribution. By collaboratively collecting data from value chain and processing it with machine
learning models enables the detection of trends that links demand management, order
management, order fulfillment to logistics carriers and distribution centers leading to improved
ability to react to changes in value chain and avoid costly situations. One example of this is the
ability of such solutions to greatly simplify Omni Channel distribution.
Cloud SaaS Advanced
Analytical solutions are
the basis of creating the
Factories of the Future
today and Omni Channel
Retail Supply Chains
8. Characteristicsofsolutionsfor
ValueChainBusinessTransformation
8 Cloud, SaaS and Advanced Analytics for Value Chain Business
Transformation
Here are some key characteristics of transformational value
chain solutions.
Almost everyone of the solutions that I have been involved in
that have delivered innovation and transformational business
value has taken the crawl, walk and run approach.
Innovation requires out of the box thinking and also
challenging the norm and status quo. It is important to build
credibility in the key stakeholders to remove skepticism that
generally exists. It is important to also prove that the solution
meets its stated goals. The best way to do this is to pick a
important but tangible out-come to focus on as the initial scope of
the innovative solution and focus on accomplishing this. This is the
crawl part. Although we are crawling it is also very important to
keep this phase relatively short term in nature. Example of this is
when proving out the use of machine learning models for demand
forecasting, it may make sense to pick one product family where
issues have been observed with forecast accuracy or rapid demand
changes occurring. Taking the results of this proof of concept
phase lays the confidence building foundation to plan the
transition to most of the products thus leading to walk and run
phases.
Best practices on machine learning models recommends
usage of proving the model locally first by using training data and
then only moving it for the actual production usage. Taking the
same earlier example use actual data from selected product
families to train the machine learning algorithm, check the results
of the demand forecast for target products in the product family
with additional sample data, review the recommendations and
then once proven leverage the model for rest of the entire data set
Innovative
solutions require
crawl, walk and
run approach to
build credibility by
taking the
important crawl
step but in a very
rapid timeframe.
Best Practices
around machine
learning adaptive
models involve
training the
models locally and
iteratively
9. Characteristicsofsolutionsfor
ValueChainBusinessTransformation
9 Cloud, SaaS and Advanced Analytics for Value Chain Business
Transformation
Solutions should make it easy and economical to collaborate
within various partners involved in the business process.
Most of the current solutions do not make it easy to collaborate
within the various partners involved unless the partners are also
using same or almost similar solutions or integrations with their
associated costs are built between the solutions.
Cloud and SaaS solutions provide a easy way to get around this
problem by providing a low cost approach to get access to selective
information that has to be shared between partners. Simple
functions like capturing or viewing data can be easily provided at
very economical cost while maintaining security. Providing the ability
to access, input and view information on any media (laptop, smart
phones, iPads and similar mobile devices) and anywhere securely
changes the way collaboration takes place. Solutions are evolving
where a Supply Chain Operations controller is reviewing shortages of
key products on exception bases and is able to within the same
application either have an email, phone or chat conversation with all
the players in the Supply Chain to manage this on business real time
basis without having to leave their current user interface.
Exception based information delivery for Demand Supply
balance is a fundamental best practice to avoid delays in responding
to Supply Chain events that typically occur in today’s Supply Chains
due to data analysis paralysis, multitude of analytical tools that do
not match, visualizations that do not give business real-time data or
solutions whose business rules are not able to keep up with the rapid
changes in business scenarios in the Value Chain. The only way to
handle these situations is by leveraging the analytical capabilities
provided by Artificial Intelligence and Machine Learning based
solutions.
Cloud SaaS
solutions provide
the ability to
access, input and
view information
on any media
securely thus
significantly
changing the way
Value Chain
partners
collaborate with
each other.
This is further
accentuated by
exception based
analytical
solutions
powered by
machine learning
Example of Company who has achieved benefits Before
Predictive Commerce they had significant operational issues due to low
forecast accuracy as well as unexpected product requests from customers.
• They went through crawl walk run approach of using machine learning
for improving E-commerce product recommendations
• Followed this up by using machine learning to improve forecast accuracy
and supply chain operations
• Results
ü 8% ROI improvement over 1 year
ü Net Revenue uplift improved by 35% in 1 year
ü Lost sales reduced by 30% in 1 year
ü Forecast accuracy improved to 92% in 1 year
10. Current Solutions miss the mark on
Cloud SaaS and Advanced Analytics
Leading Edge Innovative
Solutions are now in play
The key statement that I encountered most
of the times that I have talked with Supply Chain
leaders, Business Operations
Managers/Directors/Planners is that while most
systems promise efficient planning in reality
exception based planning management becomes
a distant dream as they and their teams spend
time mostly on managing multiple
reports/dashboards, reconciliation of information
instead of focusing on business actions that are
needed. Even if this information comes in then it
is usually not business real-time and late in the
game.
Very few solutions are truly out there that
can effectively provide this capability although it
is my expectation such solutions will come up.
Some of these provide good solutions for
point solutions like Anaplan for Demand
Management and Financial Planning and Analysis,
Apttus for Quoting.
We took a note of this and came up with our
own Cloud and SaaS solution planninginabox.com
that is designed from ground up to solve the
above as well as facilitate the following:
Ability to easily integrate within existing
architectures while managing security with
multiple means of data input.
Ensure effective collaboration by enabling
the user to collaborate via email, chat, call on
either laptop, mobile devices while being in the
same user interface.
Have pre-built adaptors to process IOE-IOT
data , and have it drive the Advanced Analytics
with leading edge visualizations.
Finally but not the least leverage artificial
intelligence and machine learning for significant
business transformation.
8 Cloud, SaaS and Advanced Analytics for Value Chain Business Transformation
ReviewofMarketPlaceSolutions
Traditional solutions provided by Oracle,
SAP, IBM and highlighted by Analytical firms
like Gartner, Forrester although are well
established are constrained by the fact that
they are predominantly architected on the
on premise model and even though there is a
push by these companies to provide the
Cloud and SaaS based versions these and are
characterized by following:
They need significant integrations to
truly leverage information across multiple
organizations.
• Except for IBM which has its Watson
product none of them have significantly
demonstrated Artificial Learning and Machine
Learning capabilities that can be leveraged to
enhance Value Chain transformation.
• Although many of them provide robust
end to end capabilities within the enterprise
looking at the current state of such solutions
within the enterprise I have observed that
most of the times it results in users struggling
to get actionable information as a result of
static business rules that are typically
embedded in such solutions.
Some of the promising enterprise
solutions such as Anaplan while eliminating
most of the above issues still have gaps in
functionalities like ability to easily provide
processed information for use by other
solutions that can implement machine
learning capability on top of its data or have
weaknesses in the solutions that help
manage the back end of Supply Chain
Management while they have mastered the
solution for front end with their Demand
Planning and Forecasting, Financial Planning
and Analysis solution as an example.
11. Value chain Best Practices have developed over a long time and for
most purposes can be characterized into the following key areas:
Plan:
- Forecasting, Collaboration and Metrics
Source:
- Sourcing Operations
Make:
- Lean and Agile Manufacturing, Technology
Deliver:
- External Collaboration, Delivery Scheduling, Forward and Reverse Logistics
Fundamentals of Value Chain Best Practices conceptually do
not change, what changes is how technology advances can be
leveraged to achieve the same.
Cloud SaaS solutions provide the ability to access, input and
view information on any media securely thus significantly changing
the way Value Chain partners collaborate with each other. This is
further accentuated by exception based analytical solutions powered
by machine learning. This provides an opportunity to drive significant
business transformation in the Value Chain.
The need for this transition is crucial but it not easy. Many
complex offerings have not yet made that transition (Oracle, SAP) and
while many are emerging (Anaplan, Apttus) these do not cover all the
gaps which has led Pluto7 to offer our own Cloud based subscription
model of Supply Chain Analytics solution www.planninginabox.com
which addresses some of the business pains highlighted in this
paper.
Subscribe to our Blog : http://blog.pluto7.com
Summary
12. Following are the articles that have been referenced by the author
when creating this content
• Top 10 Supply Chain Best Practices – Multiple articles, Author’s experience, Pluto7 team
experience
• Forecasting best practices by BetterVu
• Machine Learning - A Giant Leap for Supply Chain Forecasting –Patrick Smith in 2015
• Democratizing Data Science through Automation by Data Informed
• Application of machine learning techniques for supply chain demand forecasting – Article
by Real Carbonneau, Kevin Laframboise, Rustam Vahidov, Concordia University
• Ultimate guide to machine learning by Apttus
• The Anaplan Platform explained by Anaplan
References