VOL 5 NO 3

Jul-Sept 2007

SUPPLY CHAIN MANAGEMENT:

NEW TRENDS AND STRATEGIES

SETLabs Briefings
Advisory Board
Aveejeet Palit
Principal Solutions Manager,
System Integration Practice
Gaurav Rastogi
Associate Vice President,
Global Sales Effectiveness
George Eby Mathew
Principal Researcher,
Software Engineering &
Technology Labs
Kochikar V P
Associate Vice President,
Education & Research Unit
Raj Joshi
Managing Director,
Infosys Consulting Inc.
Rajiv Narvekar
Senior Researcher
Software Engineering &
Technology Labs
Ranganath M
Vice President & Head,
Domain Competency Group
Srinivas Uppaluri
Associate Vice President & Head,
Global Marketing
Subu Goparaju
Vice President & Head,
Software Engineering &
Technology Labs
Supply Chains and Firm
Performance
In the end of May this year, AMR published an annual report that identified top
25 manufacturers and retailers that exhibit superior supply chain capabilities and
performance. An analysis of AMR’s top 10 supply chain performers, which included
Nokia, Apple, P&G, Toyota, Wal-Mart, Tesco and Best Buy, revealed that these
firms stood apart because of their exceptional capabilities in supplier development,
collaborative product development, demand shaping and demand sensing
capabilities, process design, lean manufacturing and innovative retail practices.
Their supply chain leadership reflected both in their metrics related to execution
and market performance – return on assets, revenue growth, and inventory turns.
Consistently, the AMR’s supply chain top 25 as a whole outperformed the Dow Jones
Industrial Average, the S&P 500, and the NASDAQ.
On the other hand, in the list of 11 greatest supply chain disasters (published by
Supply Chain Digest in January 2006, the most comprehensive account I could find),
many of the firms filed for bankruptcy or were acquired, most lost market share
that never recovered, faced nasty PR fiascos, couldn’t fulfill thousands of promised
orders, missed critical seasonal shipments, faced frequent stock outs, inventory pile
ups, and inventory write-down and so on. And we are talking about some of the
very companies who topped this year’s high performance list – Apple (1995), Cisco
(2001), and Nike (2001). Heartening isn’t it, that companies do recover to leadership
positions from supply chain disasters.
In this issue of SETLabs Briefings, we look at a host of issues that appeals to our
readers in their journey to improve firm performance via the supply chain route. We
have looked at a critical issue of supply chain risk both in the context of a flattening
world and flexibility and provided a framework for managing supplier risk. In our
technology coverage, we look at some of the opportunities and challenges using
Grid Computing to tackle supply chain challenges. We present a case for SOA-
based integration for collaboration in automotive supply chain and an application
architecture using XQuery as a possible solution to counter data aggregation
challenges in a CPFR scenario. We offer practitioner’s insight on some of the
learnings from supply chain mistakes and experience, uncovering untapped savings
opportunities in financial supply chains and supply chain planning in semiconductor
industry. We provide insights on how enhancing internal visibility can reduce huge
transportation costs and take a closer look at procurement with a view of managing
costs better.
As always, we would love to hear from you. Do write back to us, if this issue has
been useful to you.
George Eby Mathew
george_mathew@infosys.com
Editor
 SETLabs Briefings
VOL 5 NO 3
Jul-Sept 2007

Overview: Creating Supply Chain Flexibility in the Flattening World 3

By Sandeep Kumar and Ashish Kumar Tewary, PhD
The ever flattening world is bringing with it supply chain risks. Are companies building
flexibility to manage such risks? The authors probe the existing supply chain trends.
Opinion: Inventory Optimization: A Necessity Turning to Urgency 15
By Greg Scheuffele and Anupam Kulshreshtha, PhD
Greater variability and uncertainty across global supply chains have increased the complexity of
manufacturing and distribution. Inventory optimization, backed by advanced technology, is the
only way out, feel the authors.
Financial Supply Chains: Applying Classical SCM Techniques to Tap Savings 27
By Siva Padmanabhan and G V Ganesh
Drawing from their vast supply chain experience, the authors opine that financial supply chains
are full of untapped savings opportunities. They believe that in an increasingly networked
environment, systematic and proven techniques can help realize such opportunities.
Technology: Tackling Supply Chain through Grid Computing: 35
Opportunities and Challenges
By Sumit Kumar Bose, PhD and Shubhashis Sengupta, PhD
Petrochemical industry is enswathed with challenges within the supply chain context. According
to the authors, Grid technologies offer hope to tackle petrochem supply chain challenges.
Insight: Learning from Supply Chain - Mistakes and Experience 43
By Narayanan Sampath
The author draws from his vast implementation experience to point out common mistakes
in global supply chain management and offers effective strategies to tackling the problems.
Planning: Global Trends in Supply Chain Planning in Semiconductor 53
Industry
By Arnab Banerjee
The author tracks the supply chain eco-system in the semiconductor industry and evaluates
supply chain planning scenarios in both the collaborative and non-collaborative systems.
Transportation: Can Transportation Costs Be Stemmed Even Before They 63
Occur?
By Suresh Prahlad Bharadwaj, Sumesh George and Hariharan Noorani
Huge inbound transportation costs are largely due to lack of internal visibility of
transportation needs. How does one reduce inbound transportation costs? The authors offer
a unique perspective in the way they address this question by tackling the problem at its root.
Procurement: Managing Costs by Leveraging Procurement Information 71
Intelligently
By Rajib Saha
In this research paper, the author contends that a 360° approach to looking at where
enterprise spends are being made can help procurement departments manage costs better.
Perspective: Aligning Treasury Services Offerings to the Corporate 83
Financial Supply Chain
By Sabitha Vuppala and Sujata Banerjee
Innovation is the key to surge ahead in a commoditized treasury business service. The
authors probe the drivers of change and look at the emerging trends in the industry.
Framework: Management of Supplier Risks in Global Supply Chains 93
By V. S. Srividhya and Raj Jayaraman
Supplier risk management has, of late, gained prominence in the wake of low cost country
outsourcing, geo-political instability and recurrent natural disasters. The authors propose a
simple SRM framework to mitigate such risks.

Application: A Supply Chain Collaborative Solution using XQuery 105

By Bharani Shivaraja Govindasamy
What could be the best way to aggregate supply chain collaboration data that lies in disparate
data sources? The author indicates at an application architecture using XQuery as a possible
solution to counter data aggregation challenges in a CPFR scenario.
Integration: SOA Based Integration for Creation of Collaborative Supply 113
Chain in Automotive Domain
By Krishnendu Kunti, Terance Dias and Ashwini Jeksani
In an agile world, traditional means of point-to-point integration of collaboration in automotive
supply chain is not feasible. The authors make a case for SOA-based integration platform.
“Supply chains can be shrunk through an efficient movement
of resources along the three classic dimensions of the supply
chain — inventory, information and money.”

Dr. Badri Devalla

Senior Principal Consultant
High Tech and Discrete Manufacturing Unit
Infosys Technologies Limited

“Enterprises that engage 360° approach in their cost

management understand their ‘as-is’ procurement state….
and identify opportunities to reduce costs.”

Rajib Saha
Senior Consultant, Manufacturing &
Supply Chain Practice
Domain Competency Group,
Infosys Technologies Limited

4
 SETLabs Briefings
VOL 5 NO 3
Jul-Sept 2007
Creating Supply Chain Flexibility
in the Flattening World
By Sandeep Kumar and Ashish Kumar Tewary
Counter supply chain disruptions by building
solid ‘sense and respond’ capabilities
M odern day business is beset
changing governance paradigms. The
with
recent surge in business literature has veered the
way corporations are getting flatter and flatter.
Flat world has emerged as the mantra by which
businesses are getting redefined.
Flat world encompasses an environment
where traditional barriers to trade are breaking
down; hierarchies – both organizational and
customer are being flattened; information and
knowledge asymmetries across competition,
customers and within the enterprise are being
eliminated. The competitive playing field is
being leveled.
EMERGENCE OF FLAT SUPPLY CHAINS
The forces of globalization and the ever
flattening world are exerting renewed pressure
on global supply chains. Supply chain flexibility
has moved beyond managing quality, cost and
on-time delivery for a variety of norms and
tolerances. It has come to include new product
introduction cycle, partner integration and
3
quality of supply chain collaboration, as also
the process and functional modularization of
supply chains.
Cost pressures, dearth of talent pools
and emerging markets were some of the reasons
why supply chains assumed global dispersion.
Following Thomas Friedman’s expostulation of
flat world, researchers at Infosys have identified
four operational shifts that are driving mega
transformation in the way demand is being
created for products and services globally and
the way it is being fulfilled [1]. We try to probe
these shifts in the supply chain context and look
at how flexibility can be achieved in supply
chains in this flattening world.
First and foremost, companies in their
journey to being the China price, are continually
looking to refine their cost structures. This
journey has taken many a company eastward
with their sourcing strategies. However it is not
only about reducing costs but also about entering
new market segments by virtue of being able to
sell at lower prices, while making money.
 As an example, GE Medical Systems
(GEMS) redesigned its product offerings to
offer 80% functionalities at 50% the US price.
This redesign was done at its China R&D
centers. GE Medical Systems is now the market
leader in China for these products and it is also
selling these products in price-sensitive market
segments in the US. Another example is that of
Cisco which funded its investment in R&D and
sales expansion by releasing over $2bn from
Selling, General and Administrative Expenses
(SG&A) globalization by reducing SG&A as
a percent of sales from 44.8% in FY01 to 36.3%
Companies stand to win the innovation game by
exploiting their global talent pools in co-creating
revenue-influencing innovation models
in FY05. For any new initiative, globalization is
always in consideration, and Asia is always on
the table [2].
A second operational shift is in the
way companies are looking at the problem
of customer loyalty creation and retention. It
is the product and service innovation cycles
today that are determinants of better customer
experience.
Today’s supply chains need to have the
required integration to help manage demand
percolation seamlessly across supply chain
partners. With customers being more and more
integrated into the product and service creation
process, customer experience calls for greater
transparency and visibility into the entire order
fulfillment cycle.
4
Change at a rudimentary level can be
observed today, in that, the demand and the
supply chain are so tightly linked that the product
or service design concept is well integrated with
the supply chain. As an example of companies
leveraging global talent and co-creation models
for rapid product innovation, Eli Lilly launched
“InnoCentive”, an online community based
business model to access global talent pool [3].
Eli Lily and other pharmaceuticals/chemicals
companies can “post” R&D problems on the
portal “InnoCentive” (with associated award
amounts) that can be solved by any of their
scientists around the globe. P&G’s strategy of
open innovation __with customers, partners and
other outside sources - now produces over 35%
of the company’s innovation, thereby influencing
billions of dollars in revenue. With over 35% of
new products having co-creation elements from
outside P&G, R&D productivity has increased by
nearly 60% in 2006 and has helped launch more
than 100 new products with significant outside
participation in the last 2 years. Significantly,
R&D costs have decreased to 3.4% of sales from
4.8% in 2006 [4].
Making money from information – the
third operational shift - has always been intrinsic
to supply chain management. At the core of all
successful supply chains over the years, has
been, excellence in information management __
both in terms of creating a robust information
backbone underlying all physical and financial
flows as well as monetizing this information
for decision support and business intelligence.
As an example, using a sophisticated demand
driven supply networks (DDSN) strategy while
leveraging an information architecture that
marries sophisticated planning, fulfillment, event
management and supply chain integration, Dell
pioneered its famous direct distribution model
that enabled it to grow its global PC market
share from less than 3% in 1995 to over 18%
in 2005 [5]. Likewise, Amazon.com embodies
In today’s dynamic business environment, only
mutated supply chain models stand to absorb global
supply chain risks
the use of advanced multi channel commerce
capabilities to efficiently fulfill millions of multi-
item orders through seamless information flow
and automation __ a feat that requires very
sophisticated coordination and timing [6].
The operational shift on “winning in the
turns” underlines the importance of being able to
manage business cycles through business’ crests
and troughs. These business cycles typically
call for a close look at existing business models
and predicate disruptive changes. Supply chain
risk management has become a key theme in
managing such disruptions successfully. Both
operational risk as well as event risk need
to be addressed squarely through predictive
mechanisms as well as through business proofing
for managing scale and scope flexibility. As an
5
example, Intel timed its asset acquisitions during
business downturns to manage rapid up-scaling
in the boom years. By sharply accelerating
spending during the 2001 recession, Intel built
up additional manufacturing capacity. During
2002-2003 recovery, Intel was able to quickly
and successfully launch new products, months
ahead of schedule. In 2003, Intel reported its
highest rate of growth since 1996 and increase in
net income by 81% [7].
From the discussion above, we
understand that the forces of globalization are
leading to a mutation of the supply chain models
in the rapidly changing world. Supply chains are
becoming increasingly global and flatter. While
global sourcing and targeting emerging markets
has become fundamental to every corporate
strategy, it is also leading to emergence of virtual
supply chains with complete supply chain
functions being outsourced. The bottom-line is
a greater propensity of risk in terms of supply
chain management globally.
SO HOW ARE SUPPLY CHAINS REACTING
TO THESE FLAT WORLD FORCES?
Empirical evidence suggests that supply chains
are morphing to help deliver to these flattening
forces. Multiple supply chain structures are
emerging to successfully address changes
caused by flattening forces. Emergence of
loosely coupled global supply chains,
redefinition of “core”, and increasing use of
counter-trade in cross-country businesses are
some of the key changes.
Value chain complexity is on the rise with
supply chains becoming lengthier. As more links
get added and niche players emerge, the moot
question is on the need for command and control
capabilities that allow companies like Cisco, a
strong proponent of outsourcing, to manage a
supply network that has more than 300 suppliers
[8]. While companies have effected supplier
consolidation and rationalization to manage the
expanding supplier base, the move towards third
party sourcing is making the average length of
supply chains longer.
Emergence of niche supply chain entities - As
roles within the value chain get blurred, new
entities are emerging that help manage global
dispersion and spread of business operations
as a core competency. In the process they also
expand their role in the value chain. An example
in the automotive industry is that of Magna,
which aims to provide complete vehicle, design,
engineering and assembly services to its OEM
customers [9].
Globalized supply chains – Supply chains are
becoming more and more global with supply
chain functions being physically distributed and
dispersed. For example, Boeing 787 program
transforms its global supply base spread across
countries as dispersed as the US, Japan, Italy and
Taipei into design and manufacturing partners.
To beat Airbus, Boeing has deployed a radical
strategy where more than 100 design partners
collaborate on the design and engineering
and more than 130 structural and systems
suppliers synchronize their operations to achieve
6
final assembly in Everett, Washington [10]. A
thought here is that globalization and JIT do not
seem to go well together as we are gravitating
towards stronger and centralized planning
functions.
Demise of “core”- There is nothing called
“core” anymore. The term “core” is relative
and it’s about supply chain modularization
today. There are companies that are outsourcing
complete functions that used to be called core in
the yesteryears.
Emergence of intricate supply chain networks -
Companies are moving away from the
integrated supply chain concept and today we
see loosely coupled supply networks. SLA-
based transactional relationships rule. Long
term contracts between partners do not always
work. Take a look at the high technology
contract manufacturers who are trying to work
themselves into shorter and more flexible
contracts to help maintain their profitability.
Another interesting aside is that companies
today collaborate in one geography and
compete in another. Hutch Essar and Bharti
compete independently, but they both have
entered into an MOU relating to a comprehensive
range of infrastructure sharing options in
India [11].
Technology is no longer the bottleneck - The
single biggest disruptive influence on business
has been technology. As communication costs
go down to virtually nothing, processing
power keeps following Moore’s Law and
use of technology becomes ubiquitous,
many companies and economies are doing
leapfrogs skipping many an evolutionary step.
Retailers used to models such as re-order point
based supply chain planning (primarily as a
consequence of the number of stock keeping
units as well as the shorter planning buckets) are
switching to time-phased planning techniques,
as earlier constraints around processing power
and processor costs have been sorted out.
Technology has also made possible the concept
of work modularization in the connected world
where work packets can be delivered virtually
out of anywhere and at any time.
The G Factor – As global trade barriers keep
falling down, governments everywhere are
putting in counter trade agreements to stimulate
local industry further fuelling globalization
of trade. Another aspect of the governmental
factor is the regulatory norms that governments
prescribe and the impact that it has on the
way businesses operate. Protocols and
regulations such as the Kyoto protocol and the
ROHS (Reduction of Hazardous Substances)
significantly impact certain businesses in certain
regions.
HOW DO WE QUALIFY THE NATURE OF
SUPPLY CHAIN RISK?
The forces that are flattening the world have
created many supply chain risks in addition to the
oft spoken benefits. With supply chains literally
disintegrating, product designers, marketers
and manufacturers that were previously housed
in a single facility are now spread over several
continents in organizations with different
cultures, languages and business objectives.
These changes have brought new risks and
challenges. Long-standing challenges, such as
short product lives and uncertain demand, have
become even more vexing.
When it comes to global supply
chains, the potential for disruption comes in
many forms, from large-scale natural disasters
and terrorist attacks to plant manufacturing
7
fires, wide-spread electrical blackouts and
operational challenges such as shipping
ports too small to handle the flow of goods
coming into a country. Today’s leaner, just-
in-time globalized supply chains are more
vulnerable than ever before to natural and
man-made disasters __ a reality that creates
greater demands on companies to keep supply
chains flexible and integrate disruption risk
management into every facet of supply chain
operations. The reason is undoubtedly that,
with longer paths and shorter clock speeds,
there are more opportunities for disruption
and a smaller margin for error if a disruption
takes place.
We identify two main sources of supply chain
disruption risk:
• Operational supply chain risk : Abrupt
discontinuity of supply (when a main
supplier goes out of business), people
(labor strikes, talent shortages), process
(internal process changes initiated
through external factors or through
internal process improvement measures),
systems (system failures, data security)
and procedures fraud
• Business risk: Macro business risks
such as market risks, financial risks,
regulatory risks (stringent emission norms
driving new engine/ manufacturing
process technology), socio-political
environmental risks, natural hazards
such as earthquakes, hurricanes, storms
and risks emerging from terrorism and
political insurgencies.
The essence of risk management boils down to
adequately appreciating the risks that a company
is exposed to for different areas of business;
identifying the ‘choke points’ along the supply
chain that would completely harm a business if
disruption occurred; and then taking the right
set of preventive measures to allow for some
protection, remembering to periodically review
your supply chain plans and risk assessment
priorities.
ARE COMPANIES BUILDING THE
FLEXIBILITY TO MANAGE FLAT WORLD
SUPPLY CHAIN RISKS?
We define supply chain flexibility as the ability to
reconfigure the supply chain, altering the supply
of product in line with customer demand. Flat
world forces are changing the face of the global
supply chain along with the flexibility that is
needed to reap the benefit from globalization.
In the flat world, companies that have solid ‘sense and
respond’ capabilities tend to amass flexibility in their
supply chains
Companies’ contingency planning to deal with
eventualities and ability to reconfigure the
supply network and product design at low cost
to meet newer market needs are having larger
impact than ever before on the supply chain
flexibility.
In this context of flat world forces,
supply chain flexibility is about having the
“sense and respond” capabilities to detect the
changes early on and having the right supply
chain structure, processes and product design in
place to respond in a timely and efficient manner
to the changes. By making it easier and less
expensive to change the source of supply, firm
8
gains flexibility to quickly realign the supply/
demand mix to satisfy the changing global
demand.
However, early detection of changes,
demands a well structured global planning
and event monitoring mechanism for global
coordination across functions and partners.
Globally integrated information systems are
critical to reduce the cost of communications and
to make relevant information readily accessible.
Global supply chain flexibility is then
all about connectivity and global business
optimality – source resources and manufacture/
deliver from where it is most cost optimal and
sell where it is most profitable. Downside of this
is increasing interdependency and much more
risk. Managers of global supply chains should
realize that they are coordinating three types of
flows __ material, information and cash flows.
Political, technological or natural events could
leave organizations isolated and exposed to
shortages of material. Companies that sell in the
United States, but have substantial portions of
their supply chains in China or in other countries
with currencies likely to appreciate against
the dollar, face a significant risk of mismatch
in their expected U.S. revenues and non-U.S.
costs. Bankruptcy of a key supplier without
any prior notification can stop assembly lines
of an OEM.
 Problems can range from dock strike in
California, a tsunami in Asia, hurricane in New
Orleans to physical damage at a supplier that can
result in catastrophic impact for a company. One
example is a fire at a single-source supplier for
Ericsson, which led to lost sales of $400 million
dollars accompanied by a drop in the stock price
by 1%, culminating in Ericsson’s exit from that
part of the business [11]. Even political problems
can bring about supply chain disruptions. Take
the case where new agreement between the
European Union and China caused a limit on
annual imports. This in turn caused 80mn packs
of clothing impounded at EU ports and borders
even though retailers had ordered their autumn
stock well before this agreement was enforced.
detrimental to functioning of supply chain
connections. Companies should act on two
overarching strategies of (i) building supply
chain responsiveness, and (ii) building visibility
to supply chain information.
Action 1: Build Supply Chain Responsiveness
through design for desired flexibility in the
supply chain structure, supply chain processes
and product design.
I) Supply chain structure:
Supply chain structure is the network
of organizations that manufacture and
deliver products or services from the
source to the customers. It deals with

Flexibility coupled with visibility to information

helps companies negotiate supply chain issues
with utmost ease

The leverage on globalization is supply network structure, the human

tempered by the management overheads of resource practices and workforce
supply chain connections. It is important to capabilities.
ensure that all the connections are working 1) Supply network should have the
uninterrupted to eliminate any risk to the ability to add and remove suppliers,
stable functioning of the supply chain. select suppliers who can add new
However, companies need to do much more as products quickly, vary supplier
Aberdeen Global supply chain benchmark relationships and have suppliers make
report in 2006 found that only 11% are volume changes. Also supply
actively managing the supply chain disruption network should be flexible
risk [12]. to adapt to new security requirements
Availability of information at right and provide guidance to its partners
time in combination with flexibility can in order to comply with new
enable organizations to respond to changes regulations.

9
 2) Supply network should provide the the ability to change processes as
ability to companies in transferring demand changes, and also the ability
production from one plant to another to adjust capacity.
when capacity is constrained. 2) Distribution flexibility: Logistics
3) Companies’ culture should allow processes should have the ability
building integral cross functional to adjust to global requirements,
teams to prevent decisions based serve distinct customer shipping
on local optimization. Hewlett requirements, vary warehouse space,
Packard has adopted this concept vary transportation carriers, and
for many of its products such as the introduce product postponement.
Desk Jet printer, even going to the 3) Ensure alternate sourcing strategies:
lengths of re-designing it so that a Alternate suppliers should be truly
generic semi-finished global version divorced from the risks borne by the
could be built centrally with preferred counterparts. This portfolio
localization being performed by of suppliers should also include
regional partners[13]. transportation providers. During
4) Workforce capabilities should the launch of the highly anticipated
be looked from a holistic perspective Fusion, Ford manufacturing in Mexico
to encourage involvement of was hamstrung by the bankruptcy of
specialists. By working closely with Collins & Aikman, a primary parts
specialist providers, greater levels of supplier. Ford, along with other
customer value can often be automobile manufacturers, was
achieved at lesser cost. Auto makers compelled to provide financial
use 3PL specialist logistics providers, backing to the floundering parts
where a 3PL runs the warehouse supplier to ensure continuity in
and OEM takes ownership of manufacturing operations [13].
inventory only when a production Companies such as Dell and Cisco
line calls for it. have suppliers in different continents
for added flexibility.
II) Supply Chain Processes 4) Combine operational and financial
Supply chain practices should have hedging: BMW after frequently
contingency plans to deal with all kinds suffering from strong Euro
of eventualities. It is almost impossible to appreciation in its heavily European
predict all the eventualities in a global based production operations, has
business environment. Elements of employed a combined hedging
supply chain should have flexibility to strategy[14]. It has announced
quickly respond to unexpected changes. expansion of production and sourcing
1) Systems and process design for facilities in the North American and
flexibility: Operations systems Asian continents in an effort to create
should have the ability to reconfigure a more globally diversified supply
assets in line with customer needs; chain, thus adopting more integrated

10
 risk management approach in its Action 2: Build Visibility to supply chain
handling of global risks. information as this is the key to detect
5) Ensuring business continuity: and manage change. The company and its
Business continuity plan should trading partners and their partners need to
include a wide range of contingencies, be very much a part of the risk-management
disaster recovery, the safety of process. Visibility helps in getting a pulse
employees, the retrieval of backup of the environment in which supply chain
business data emergency is functioning. This way it not only helps in
communication, possible relocation early detection of changes but also enables
of business operations and the collaboration amongst partners.
sourcing of goods from alternate I) Ensure visibility to manage change:
suppliers. Integrated systems to coordinate activities
III) Product Design throughout the entire supply chain from
Product design should be modular so as raw materials and components to the
to reconfigure the products to minimize end-consumer.
the risks in meeting market demands at 1) Improve visibility of in-transit
lowest cost. This can be achieved by activity: Incorporate visibility tools
product design that enables component to both track product movements
commonality across products and and also obtain information with
postponement of final product which to make financial decisions,
configuration. serve customers, respond quickly to
1) Commonality is a big focus in Ford. competitive threats, cut costs and
Commonality of parts across existing speed delivery.
and proposed vehicle lines reduces 2) Improve visibility of supplier
costs and leverages economics of scale. activity: Global supply chains are
This allows aggregation of demand characterized by interdependencies.
and minimizes the inventory buffer It is important to be signaled about
to address a particular risk level. partners’ delivery performance and
2) Postponement is about delaying their compliance with regulations.
final configuration of an item for as Lead indicators should be agreed
much duration as possible. The aim upon and monitored to have
of the global supply chain should visibility into supplier activities.
be to carry inventory in a generic 3) Improve end-to-end automation:
form, i.e., standard semi-finished Provide enhanced insight into current
products that are awaiting final and time-phased end-to-end inventory
assembly or localization for enhanced positions – in motion and at rest,
responsiveness. For example, Dell including vendor managed inventory
assembles different computer – as well as mobile assets such as
configurations based on market containers and equipment. A key
demands. HP also uses the same future enabler for this is RFID
concept for its Desk Jet printers [15]. technology.

11
 4) Create visibility into lead indicators: in an outsourced business model.
Business intelligence (BI) and This cultural buy-in and visibility
affiliated data warehouses and event- on both sides must be demonstrated
management tools track events and in day-to-day behavior at all levels,
exceptions in supply chain or there will be cultural polarization
performance by constantly monitoring between the business and the outsourced
leading indicators to supply chain delivery function.
problems before they occur. BI has Toyota was able to manage the risk
the sifting and sorting capability to caused by bankruptcy of one of its
identify suppliers, routes, carriers and key supplier. It applies weekly get-
ports that threaten business together for managers over videocon-
continuity. ference to discuss any new rumors
5) Improve visibility of trade and potential risks to work out a recovery
agreements for origin management: plan just in case.
Companies can bring down the cost of Enterprise solution providers such as
their products by better automation SAP have a number of products in the market
of origin management programs and that are helping companies in developing
understanding on how to use sense and respond capabilities – in automatic
preferential trade agreements in data acquisition and transaction processing
their product design, sourcing for supply chain automation through Auto-id
and distribution decisions. Renault infrastructure (AII); in setting and monitoring
has used origin management lead indicators through Event management
information housed in its centralized (EM); in identifying links, products, partners
trade compliance database to drive a in the supply chain that are prone to higher
whole new low-cost car. risk through BI/BW; in having a flexible IT
6) Improve visibility of regulation architecture to have plug and play capabilities
compliance: Globally spread supply and seamless data exchange with partners
chain is highly vulnerable to through service oriented architecture. These
penalties caused by violation application scenarios are supported by having
of country specific regulations a “single version of truth” achieved through
regarding environment, safety, labor common data definition across the supply chain
etc. SOX compliance is one way that enabled by master data management (MDM).
can help top management in managing
this risk. CONCLUSION
II) Ensure collaboration to manage changes Globalization and the forces of the flattening
Automakers try to give supplier advance world are creating huge supply chain risks and
notification of production changes. GM it is unlikely that these would fade in the
uses its portal “GMP SupplyPower” web near future. Cost considerations are driving
o
site to post updates for its 360 global companies to explore innovative sourcing
suppliers. It is also important strategies for physical goods as well as services.
to achieve cultural alignment especially The pressure on refining costs of doing business

12
and the consequent move to tap resources audit process is essential to providing on-going
globally is exposing companies to greater feedback to management and supply chain
supply chain risk. An effective risk management participants on the performance of their facilities
framework ensures that companies not only and their compliance with agreed, supply-chain
manage their supply chain disruptions better wide standards. Organizations can build a solid
in these scenarios but also put in early warning sense and respond capabilities in their global
systems that help identify these upfront. supply chain using enterprise products. How
There will always be natural disasters, organizations manage their enterprise risk will
as well as corporate mistakes. In order to insulate predicate success and failure and determine
themselves from the consequences, companies whether they are able to win in the turns when
may have to spread their risks more widely. supply chain disruptions occur.
The cost quantification of supply chain disruption
risk could potentially change the significance of REFERENCES
the cost arbitrage considerations that drive most 1. www.thinkflat.com.
global sourcing decisions. 2. Jeff Sanford, Beat China on Cost, Canadian
That does not necessarily mean fewer Business, November 2005.
aircrafts will be queuing up to land at Louisville 3. www.innocentive.com.
and Memphis, or that fewer container ships will 4. P&G’s New Innovation Model, Harvard
set sail from Asia’s bustling ports. But it could Business Review, Vol. 84, No. 3, March
mean that in future companies may spend much 2006.
more to maintain a number of different supply 5. Tony Friscia, Kevin O’Marah and
chains based on the risk assessment and some of Joe Souza, The AMR Research Supply
those may be closer to home. Chain Top 25 for 2005, Nov. 8, 2005.
Information driven command and 6. Filling Amazon’s Tall Orders,
control capabilities will assume critical Business Week, December 2005.
significance to help manage integrated but 7. Peter Navarro, The Well Timed Strategy:
modularized supply chains. A formal risk Managing the Business Cycle for
assessment mechanism is needed to identify Competitive Advantage, Wharton School
key processes that are likely to be affected by Publishing, Mar 2006; and Darrel
disruptions and characterize the facilities, assets Rigby, Moving Upward in a Downturn,
and human populations that may be affected. Harvard Business Review, Jun 2001.
Traditional risk management approaches 8. Stephen Hochman, Value Chain
should then be undertaken for each key process Complexity, Part 1: What It Is, Why It
to identify vulnerabilities, triggers for these Matters, AMR Research, Mar 1, 2007.
vulnerabilities, likelihood of occurrence and 9. Annual Information Form, 2005, Magna
mitigation and risk transfer activities. Reporting, International (http://www.magna.com/
periodic auditing, management and legal magna/en/investors/governance/
reviews of implementation plans and on-going documents/pdf/AIF.pdf).
results (e.g., of near-miss management and 10. Lora Cecere, The Boeing 787: Demand-
other disruption risks) complete the business Driven Strategies Take Flight, AMR
process for disruption risk management. The Research, May 4, 2006.

13
11. What Vodafone will collect from the Hutch
call, The Hindu Business Line, Feb. 18, 2007.
12. Global Supply Chain Benchmark Report,
Aberdeen Group, 2006.
13. The Detroit News (http://www.detnews.
com/2005/autosinsider/0510/20/C01-
14
355079.htm).
14. http://news-info.wustl.edu/news/
page/normal/5446.html.
15. Ting Shen, A framework fro developing
postponement strategies, MIT Research
Paper, February 2005.
 SETLabs Briefings
VOL 5 NO 3
Jul-Sept 2007
Inventory Optimization:
A Necessity Turning to Urgency
By Greg Scheuffele and Anupam Kulshreshtha
Counter your supply chain uncertainties
with inventory optimization techniques
and technologies
A key concern for global manufacturers
today is reducing inventory and inventory
driven costs across their supply and distribution
networks [1]. Pressure to cut inventories continues
to build for several reasons. Manufacturers no
longer manage linear or stable supply chains.
They juggle vast supply networks. Globalization
of the supply network and supply base drive
higher inventories and make cutting inventory
more difficult. Globalization among consumers
is putting pressure on product availability,
prompting manufacturers, distributors and
retailers to upgrade their stock keeping
policies. Emerging market consumers are
becoming as demanding as those in developed
markets. These challenges are exacerbated
by manufacturers’ own product development
decisions. The drive to innovate and increase
the rate of new product introductions leads to
high rates of new technology adoption for next
generation products, putting enormous pressure
15
on inventory management across extended
supply chains.
In this context, manufacturers have
difficulty reducing inventory with traditional
or even advanced inventory management
techniques. Today’s global manufacturers have
largely hit limitations in leveraging material
requirements planning and management
processes and systems to cut inventories. Even
advanced inventory management techniques,
such as sales and operations planning or
developing demand-pull replenishment systems
with suppliers using Kanbans, have been either
embraced or found to deliver less impact on
overall inventory reduction than anticipated.
In the last few years a new paradigm
has emerged: where one finds operations teams
and planning teams of the manufacturer applying
the latest techniques and technologies to improve
inventory visibility, control and management
across the extended supply network.
 Uncertainty
Hedge through
increased inventory
Figure 1: Addressing Uncertainty through IO Techniques
We call this collection of efforts as
Inventory optimization. Inventory optimization
helps discrete manufacturers control inventory
driven costs and address today’s demand
volatility and supply chain complexity.
We discuss several trends in inventory
optimization including opportunities to apply
this new technology to solve more that just high
inventory costs. Inventory optimization can
enable smarter product launches, lower direct
material cost of goods and faster manufacturing
and distribution execution.
NEED FOR OPTIMIZING INVENTORIES
There are several reasons manufacturers are
increasing focus on optimizing inventory by
applying the latest tools and techniques for
inventory control. Traditionally, competitive
pressure has always driven manufacturers to
seek enhanced capabilities to reduce inventory
16
Reduce through
forecast accuracy
Address head-on
through flexibility
Source: Infosys Research
levels; to enhance service levels and supply
availability; and to establish the right product
inventory mix and level in each geography and
channel. Many manufacturers also focus on
inventory as part of shifting their operations
to achieve demand-pull replenishment across
their supply network – hoping to achieve
the performance demonstrated by leading
manufacturers who have succeeded in this such
as Dell, Procter and Gamble, Nokia, and Toyota
Motor [2].
A key driver of the renewed focus on
inventory lies in the recognition that traditional
techniques are failing to reign in inventories in
the wake of increased supply chain complexity.
This complexity is characterized by increased
uncertainty. Demand is more volatile and
therefore less predictable. This is true not
only for aggregate demand but for forecasting
splits and volumes across channels and
markets. Traditionally, three strategies have
been employed by manufacturers to address
uncertainty: (a) increase inventory levels to
hedge against uncertainty, (b) develop supply
chain flexibility to be more responsive to
uncertainty, (c) improve forecast accuracy so that
less uncertainty propagates to the manufacturing
floor. Inventory optimization techniques and
technologies map to the flexibility and accuracy
strategies [Fig. 1].
What is driving the dramatic increase
in the complexity (and therefore uncertainty)
of managing large supply and distribution
networks? Globalization is one big driver, the
evolution of emerging markets such as China
and India present new challenges in effective
product distribution with low inventory levels.
Globalization of supply networks means that
key functions such as R&D, product design and
manufacturing are now geographically spread
out, which hampers inventory reduction efforts,
that are often best executed by a cross functional
team working together very closely. Increased
rates of new product introduction and product
innovation are also driving complexity into
supply networks. Finally, because increases in
transportation and logistics options have made
careful control and planning of in-transit or
pipeline inventory difficult, manufacturers are
tending instead to let inventory drift upwards.
INVENTORY OPTIMIZATION DEFINED
Inventory Optimization (IO) is the application
of a range of latest techniques and technologies
for improving inventory visibility, control,
and management across an extended supply
network. As we will illustrate later in this paper,
these techniques and technologies are driving
improvements beyond what traditional inventory
management techniques – even advanced
techniques – have been able to deliver.
17
IO Techniques
IO techniques apply rigorous and discrete
analysis to analyzing inventory performance.
They then use the analysis to identify product-
specific changes to inventory stocking and
replenishment policies; to identify the supply
network configuration; or, to correlate
inventory investments to item revenue or profit
generation.
On the planning side, a key inventory
optimization technique is profit-driven analysis,
where the profit each individual product
contributes is ranked; a pareto distribution is
developed to separate high profit products from
lower profit products; and inventory holding
policies are adjusted to cut inventory on low
ranked products and increase inventory on high
ranked products, resulting in an intelligently
applied net inventory reduction.
On the execution side, manufacturers
are striving for IO by applying improvement
concepts based on lean principles, and by
expanding the use of collaborative and demand-
pull replenishment schemes such as vendor- or
supplier-managed inventory to drive highly
precise replenishment and fulfillment activity.
These techniques are also benefiting from
improved supply chain planning and control.
Lean seeks to optimize inventory by driving out
non-value added inventory management tasks
in the factory or warehouse and by improving
planning and control at a granular level across
each manufacturing or distribution step. Vendor-
or supplier- managed inventory schemes seek to
share risk and offload inventory ownership.
IO Technologies
A key inventory optimization technology is the
IO engine. IO engines reveal opportunities to cut
inventory by analyzing inventory performance
holistically - looking at data from across the
 “Classic” Inventory “Advanced” Inventory Inventory Advantage of Inventory
Management Management Optimization Optimization over prior methods

Material Constraint based IO Engine Better characterizes demand

Requirements planning (APS) uncertainty and lead time
Planning (MRP) variability
Advanced modeling
Integrates with MRP and APS

Days of Supply rules ABC Classification Profit-Driven Rationalizes inventory with

for setting inventory Analysis minimal impact on revenue /profit
levels

Cycle counting Materials Closed loop planning
Management system and analytics with
inventory control via
exception management
Provides more predictable control
over material flows
Enables faster re-configuration of
supply chain
Supports smoother absorption and
handling of unexpected supply or
demand swings impact

No control over Resorts to Chase Optimizes considering Better synchronization between

production Scheduling techniques production and Production and dispatch
transportation batches

Table 1: Different Types of Inventory Management and Source: Infosys Research

Control Techniques

extended supply network. They integrate with
Advanced Planning and Scheduling systems
and Material Requirements Planning systems to
incorporate policy updates into the supply chain
planning cycle. IO engines identify ‘smarter’
inventory holding rules and replenishment
policies that increase overall supply chain
planning accuracy. “Smarter” typically means
applying these policies at a more discrete level,
such as at an item /stock keeping location
combination level instead of just at the item
level.
IO engines characterize supply network
uncertainty present in a variety of specific steps
or links in manufacturing and distribution
processes using advanced mathematical
(non-linear, algorithmic, etc.) models that
are then solved to identify optimal inventory
policies, stocking locations, or quantities. The
uncertainties addressed by IO engines include:
demand uncertainty (or forecast error), cycle
time variability and replenishment lead time
variability. The output of running an IO engine
is fed back to the ERP, constraint based planning,
or other discrete planning system, adjusting
inventory policies as a finite level. IO engines
have a range of applications, from modeling and
optimizing safety stock across the supply chain
to identifying optimal re-order point sizes in
environments with highly erratic demand.
The rise in interest in IO engines is
likely linked to their increased ease of use and

18
accessibility over time. In the last few years,
a new generation of technology and related
software vendor community has developed
around IO engines [3]. The latest IO engines
bring the computational horsepower to solve
very large optimization problems quickly.
We make a distinction between
inventory optimization engines as a new
technology and a broader collection of concepts,
techniques and technologies called inventory
optimization.
In spite of the significant advantages
available from the latest IO techniques and
technologies, leveraging them effectively requires
much more data than traditional or “advanced”
inventory handling techniques. This is a valid
concern some manufacturers express when
planning to leverage these techniques. A second
concern is frequently raised around the complexity
of the calculations, formulae and mathematics
employed. For traditional techniques and even
for most advanced approaches, the computations
required are limited to simple calculations and
straightforward mathematical formulae, which
are simple to communicate and explain to teams
outside the supply chain function including
upper management. For inventory optimization
however, typically higher order calculations,
including complex equations in algebraic
or calculus forms, are used. Understanding
and deriving meaningful results from these
calculations requires a deeper mathematics
background and greater computing power.
Despite all the complexity, IO
techniques and technologies are gaining
ground and finding more and more applications
because of recent advances in information
technology and far greater computational power
available at the disposal of today’s supply
chain architects. Similar advancements on the
operations research front have also led to newer
19
ways to solve complex mathematical equations
in acceptable time durations. This progress is a
key reason for manufacturers to have applied IO
techniques across a wide range of problems in
their supply, manufacturing, and distribution
networks.
APPLICATION AREAS
IO techniques and technologies are being applied
within both supply chain planning and execution
processes. Manufacturer are using these enhanced
capabilities to cut inventory, enhance service levels
and maximize return on investments by setting the
right inventory levels in the right production lines
and distribution channels.
Application of an IO approach depends
on a deep understanding of the existing
environment around the supply chain. An
accurate knowledge of the various cost
elements of the supply chain, along with a good
understanding of existing lead time and demand
variability is required to model the process
correctly. Modeling also demands a proper and
accurate definition of the optimization objective.
Some suitable objectives can be minimization of
costs, maximization of revenue or maximization
of profit. One also needs to define business related
constraints in unambiguous terms. Examples of
such constraints are the process capacities of
various plants, demand limitations of various
customers and service time commitments
between different chain partners.
A mathematical representation of
these objectives and constraints represents the
model for the subject supply chain. Typically,
factors that drive the business constraints or
the defined optimization objective, or both are
nonlinear in nature. In addition, the optimized
value of the IO objective function is point in
time value that changes over time as component
parameters in the function change. The same
is true for the constraints as well. For example,
assume there is a constraint on total man hours
available for production. As the number of
production units increase, the man hour per
unit shall decrease at some rate, resulting in non
linearity in the constraint. Characterizing this
dynamic is difficult using traditional analysis,
inventory optimization is an ideal approach
for the complex mathematics this entails. In
addition, there is variability in demand and in
fulfillment and replenishment cycle times which
by themselves can be difficult to characterize.
Thankfully, advancements on the mathematics
and IT technology fronts have made it possible
to capture these situations in great detail and
solve them efficiently.
Non-linearity in factors that drive business constraints
or optimization objectives call for the adoption of
mathematics-intensive inventory optimization approach
Inventory optimization techniques
can be applied to specific areas across a broad
range of supply chain planning and execution
activities. The needs, constraints, participation,
process changes and benefits of each supply chain
partner will vary depending on the industry
and type of optimization problem. Some newly
emerging and proposed applications of IO
technology to specific operational issues and
supply chain problems are explained in the
following sections.
The following lists emerging areas of
applying IO technologies to address complex
supply chain and inventory issues:
20
Inventory Reduction in Plant Operations /
Assembly Lines
The production operations are subject to
various process center capacities. In typical
assembly line scenarios, the process capacities
of successive work centers interact with each
other in a complex fashion and have substantial
impact on the quantities of WIP or staged
component inventory for each such work
center. To reduce inventory at each work
center, one needs to optimize the inventory
requirements across the processes in a
holistic fashion keeping in mind the required
manufacturing throughput rate and existing
product availability /service commitments the
manufacturer must achieve.
Inventory requirement can be
optimized by balancing the assembly line
for a smooth work flow. One also needs to locate
the most critical resource or bottleneck in
the assembly line. This bottleneck defines
the maximum throughput rate through the
assembly line with minimum inventory
requirements. The decision has the potential
to influence the process batch size, transfer
batch size and the buffer capacities for each work
center. At a macro level, these decisions also
impact the lot sizes that are required to be
procured from suppliers and provided to the
next stage in the supply chain.
Inventory Reduction Across Transportation
Networks
Each stage in the supply chain has transportation
options available with different cost structures.
The choice of each transfer mode, or combination
thereof, impacts the inventory needed to be
carried in pipeline between those stages. A
simultaneous optimization of the engagement
of various transportation modes and the level
pipeline inventories can have significant impact
on the overall working capital invested in
supply chain inventory at any given time. Trade
offs are involved on two fronts while making
the transportation decisions. On one hand,
From inventory reduction in plant operations to
transportation networks, via sourcing policies to lot size
optimization — inventory optimization technologies can
address the most complex inventory issues
transportation costs are compensated against
the pipeline inventory. On the other hand,
transportation choices also impact customer
responsiveness and thus the service level
(product availability) commitments. A careful
selection of the right configuration of modes,
inventory holding in pipeline and service level
commitments can optimize and improve the
overall cost incurred in the chain.
Inventory Reduction via Changes to Sourcing
Policies
Low cost country sourcing strategies open
up several options for applying inventory
optimization. The key is identifying which
procurement decisions are significant enough
21
to impact pipeline and safety stock inventory
levels. Many procurement decisions and
activities only indirectly drive inventory
levels because decisions are made prior to
actual sourcing execution or because of the
longer cycle times associated with tactical
procurement. While performing the source
selection as a strategic initiative, little can be
done on inventory optimization as a tactical
exercise. The mix of procurement from various
sources provides significant benefit if optimized
along with the inventory requirement for each
possible configuration of procurement mix from
various suppliers.
Inventory Reduction via Lot Size Optimization
Cycle inventories can be reduced by decreasing
the lot sizes used in production and distribution
replenishment. In production, optimal lot sizes
depend on the fixed cost of forming the lots. A
problem arises when different produced units
have different optimal lot sizes for production
but share the same work center resource or
transportation resource. Considering these
limitations and the cost structure in place, the
lot sizes for different produced units can be
optimized so as to utilize the available resources
to the maximum. This concept can be applied
across multiple manufacturing work centers
and transportation resources via custom
optimization routines that look across the supply
 Application Area Approach with Inventory
Optimization
Inventory reduction Finding out the Critical
in plants / resource and following
assembly lines the bottleneck approach
Inventory Finding best configuration
reduction across of transportation modes
transportation (and costs) along with
networks required safety stocks,
utilizing service times to
optimum
Inventory reduction Finding best configuration
via changes to of sourcing options and
sourcing policies quantities with safety
stock requirements
Inventory reduction Finding joint lot sizes for a
via lot size suitable group of products
optimization sharing similar resources
and transportation
schedules
Risk Pooling Finding right size of
inventory by analyzing the
demand patterns across
geographies for group of
items
Inventory reduction Finding requirement
via common schedule and quantities of
component components common to a
planning group of products
Inventory reduction Postponing the product
via postponement differentiation to later
stages in chain by
involving customer
preferences at later
stages
Table 2: Inventory Optimization vs. Traditional Approaches
chain, producing a globally optimal result. An
effective optimization function would consider
local lot size optimality, the arrival rate of work
in-process lots from each upstream process, and
the storage or staging capacity at each resource.
Advantages over
traditional approaches
System runs at the maximum
capacity without unnecessary
inventory through out the line
Results in relatively lower
safety stocks due to lesser
lead times and proper
exploitation of available
transport modes
Better utilization of available
sourcing options. Easy to
evaluate multiple sourcing
policies
Better overall reduction in
inventory and transportation
costs
Better overall reduction in
inventory across geographies
Significant reduction in the
component inventory across
multiple finished products with
common component parts
Allows mitigation or
elimination of early WIP stage
inventories
No additional inventory
needed to handle varying
customer preferences
Source: Infosys Research
Risk Pooling
Safety stocks
over various time periods and for different
product groups can be aggregated to reduce
the total stock carried for providing a pre-
22
Incremental Value
Realized
Provides an effective tool for
assembly line operations and
a control mechanism for plant
related inventory
Reduces transportation and
distribution costs
Can increase actual product
availability level with no
net increase in pipeline
inventories
A comparatively longer term
cost reduction technique that
can simultaneously considers
procurement and holding cost
An integrated inventory
reduction approach for product
families,
Optimizes inventory with
suitable MRP data
A time tested approach,
Emerges as an efficient
technique for controlling
inventory in distribution
function
Value realized in the
procurement of components
for ATO or similar
environment,
A trusted inventory reduction
technique, casts drastic
impact on supply chain
complexity by reducing
number of products
across various geographies,
defined service level. Many companies like Dell,
HP and Amazon practically optimized their
safety stocks over different dimensions to
exploit the benefit of risk pooling. If the
demand patterns across geographies
product groups are independent, the variance
for the aggregated demand will be less than
the summation of individual variances. In
this scenario, IO techniques and technologies
are applied to identify (a)which geographies
or product groups are optimal candidates for
safety stock aggregation, (b) which inventory
holding locations are optimal for risk-pooling
from a cost standpoint, and (c)the level of
safety stock inventory to hold in the risk-pooled
location. The same explanation holds good
for aggregation over various time periods and
for different product groups as well.
Inventory Reduction via Common Component
P lanning
In a discrete manufacturing environment, a
lot of inventory is held as components across
various products. Thousands of components
required for various products have significant
commonality across products. Aggregation of
such components is another form of pooling,
with more relevance to high tech and discrete
manufacturing environment. Since the same
component is required for various products,
the demand at the component level is more
predictable and requires less safety stock as
compared to a simple addition of safety stocks
for the same component without considering
commonality.
Inventory Reduction via Postponement
Postponement refers modifying a manufacturing
process so that more of the operation is done
closer to the customer and on more of a just-in-
time basis.
This enables the organizations to check the
impact of variations in the consumer demands
for product variants, very early in the chain
and thus avoiding the need to carry safety
or stocks for different variants in the initial
stages of the supply chain. Both postponement
and component commonality push the
product differentiation towards downstream
in the chain and facilitate aggregation till later
stages.
Applying IO techniques here is
similar to the Risk Pooling application. IO can
identify (a) which products are optimal for
postponement, (b) where postponement should
take place physically for least cost, and (c) the
level of postponement inventory to carry.
EXAMPLES OF MANUFACTURERS
APPLYING INVENTORY OPTIMIZATION
These examples of manufacturers applying
inventory optimization illustrate real world
success with the concepts, technologies, and
techniques.
One major manufacturer of personal
communications devices is undertaking a
broad initiative to enable sell-side supplier
managed inventory for its major customers.
This manufacturer’s existing product
fulfillment model is based on a classic multi-tier
distribution channel. Before adopting inventory
optimization techniques, the manufacturer’s
distribution network was unsynchronized
and contained sequential layers of distribution
leading to limited visibility, excess inventory,
stock outs, cycle time delays and difficulty
establishing new retail relationships.
In some markets, the manufacturer
faced a 72% retail shelf stock-out rate. In
addition, its forecasts did not adequately account
for the potential demand if stock outs could be
addressed.
23
 To address these issues, the introduced because it could not ramp up
manufacturer, developed a supplier managed new product introduction fast enough. The
inventory (SMI) approach that eliminated manufacturer’s overall goal was enabling higher
distribution tiers, review cycles and order actual product availability levels during new
distortion and improved demand response. It product introductions. It used the following
also developed business and system models for approach to apply an IO engine to help determine
collaborative inventory forecasting with channel how to properly stage pre-build and pipeline
partners and customers, including short, medium inventories:
and long term forecasts of consumption,
replenishment and aggregate sales. • First, it identified a series of product
The manufacturer approached IO by lifecycle stages (Product launch phase,
deploying replenishment planning and execution Ramp up phase, Maturity phase,
solution to support SMI with data shared across End-of-Life phase), each with its own
customers and channel partners allowing the target service level. Demand
company to simultaneously improve planning characteristics for each phase were also
and execution processes. In addition, for each identified, viz.,
SMI replenishment process the manufacturer • Forecast accuracy
deployed, a replenishment lot size optimization • Demand growth
was performed using several months of • Demand variability (based on actual
historical data. demand)
After deploying its solution with • Demand price sensitivity
several customers, the manufacturer realized
significant benefits. With a single channel • Second, it developed a supply chain
partner and distributor in Southeast Asia model that characterized these
- the manufacturer was able to achieve: dimensions across timeframes
corresponding to the expected duration
• A $31 million increase in revenues of each lifecycle stage
• A $40 million improvement in cash flow • Finally it ran the IO engine to optimize
• Recovery of demand lost due to retail inventory target settings to incorporate
stock outs – meeting this demand meant in planning and procurement policies for
increased sales volume and market share. each product at key phases of its product
lifecycle. For the old generation product,
The second example is that of a inventory targets were adjusted for
diversified industrial products manufacturer. its end-of-life phase. For the new
The manufacturer has a range of industrial generation product, inventory targets
products that include industrial solvents and were adjusted for product launch and
machine tools. It faced complexity in managing ramp-up phase.
transition from the older generation product to
its newer generation replacement. It frequently By optimizing inventory targets and
overbuilt its old generation product for several replenishment policies in this manner, the
weeks after the new generation product was manufacturer was able to deliver minimum

24
cost and maximum distribution coverage for
both products. By allowing service levels
to vary with naturally occurring shifts in
demand variability across lifecycle stages, the
manufacturer was able to reduce inventory
by more than 18% with no impact on product
availability.
In yet another case, a power tools
manufacturer leveraged an IO engine working
along side Material Requirements Planning
(MRP) and Advanced Planning and Scheduling
(APS) systems to enable granular planning of
WIP inventory positions (location, target buffer
quantities, replenishment policies) throughout
Inventory optimization is the new problem solving
mantra for all supply chain related issues
its supply chain [5]. The IO engine determined
a globally optimal placement of inventory,
considering its cost at each stage in the supply
chain and also the service level targets and
replenishment lead times that constrain each
inventory location.
CONCLUSION
The increased complexity of manufacturing and
distribution among global manufacturers due
to greater variability and uncertainty across
the supply chain suggests that a new approach
to controlling and reducing inventory levels is
required. Pressures and trends impacting a
manufacturer’s ability to effectively manage
inventory at a global level are increasing.
Traditional methods such as accurate forecasting
25
and efforts to induce flexibility in the supply
chain are still necessary but not sufficient
to manage the growing multi-dimensional
complexity. A suggested approach is to adopt
inventory optimization concepts, techniques,
and technologies.
Inventory optimization is a powerful
problem solving approach backed by advanced
technology. The concepts, techniques, and
technologies of inventory optimization help
model, characterize, and account for supply
chain uncertainty. This uncertainty is a key
reason manufacturers maintain higher than
needed inventory levels. Inventory is a buffer
against the uncertainty related to variable
processing and replenishment lead times, erratic
demand and forecast bias or error,
The key to effectively leveraging
inventory optimization lies in viewing it as a
problem solving approach. A specific set of
constraints and parameters has to be identified
and modeled to characterize supply chain
behavior. An objective function is to be derived
from the model to isolate the parameter requiring
optimization. Finally, higher order mathematics
are to be applied to solving the function, often
aided by an IO engine or large scale computing
capacity. Using this approach, changes can to
be effected to inventory planning, inventory
stocking, replenishment, and transportation
processes. Also this could lead to defining
the underlying operational policies at a very management practices Outdated,
granular level. Aberdeen Group, March 2005.
4. Matthew Menner and Dan Harmeyer,
REFERENCES Developing Creative Supply Chain
1. Sensing and Shaping Demand Strategies in Highly Competitive
in a Consumer-driven Marketplace, Markets” Conference Proceedings,
Electronics Supply Chain Assoication presented by at the 2006 SCOR Supply
Report, 2006. Chain World North America Conference.
2. Kevin O’Marah, The Supply Chain Boston, MA USA. March 29, 2006.
Top 25: Supply Chain Leadership for the 5. Sunil Chopra and Peter Meindl, Supply
21st Century, AMR Research, Nov 9, 2006. Chain Management’ Third Edition,
3. Beth Enslow, Are Your Inventory Prentice Hall, March 2006.

26
 SETLabs Briefings
VOL 5 NO 3
Jul-Sept 2007

Applying Classical
SCM Techniques to Tap Savings
By Siva Padmanabhan and G. V. Ganesh

Financial supply chains are full of

untapped savings opportunity that can be
realized through systematic application of
proven SCM techniques

O ver the last 20 years, companies have

leveraged advanced management science
and process improvement techniques to radically
transform their physical supply chains. Side by
side, there has been the revolution in information
technology and the internet, resulting in even
larger leaps in their information supply chains.
However, little attention has been paid to
improvements in the associated financial supply
chains. While forecasts and documents are
exchanged in real time and more products are
being delivered within 24 hours of order, it still
takes 30-45 days or more to settle the associated
financial transactions, virtually unchanged from
20 years ago.
Working capital that is locked up in
anticipation of this settlement cycle is enormous.
It is estimated that the cost of working capital is
over 1% of the revenue of an average Fortune 500
company. Optimizing working capital has been
traditionally within the purview of the finance
function in a company with the result that the
focus has been on ensuring cash availability
(risk reduction) and improving interest yield
on cash. The accounts payable and accounts
receivable functions have been driven by
transaction processing efficiencies rather than by
a motive to drive improvements to the financial
supply chain.
In addition, limited attention is paid
to the integrated financial supply chain that
traverses companies in a trading partner
network. Each company operates according
to fairly static rules such as standard payment
and discount terms that pay no attention to the
relative value of financial assets and liabilities
to each trading partner. Each trading partner
looks for local optimization without regard to
optimizing financial flows across the trading
partner network.

27
Sounds familiar? This is fairly close to the way
product supply chains operated 20 years back.
GOING BEYOND TRANSACTION
AUTOMATION
The real time information environment in
today’s supply chains creates tremendous
opportunities to optimize the financial supply
chain without an attendant increase in operating
risk. Although the importance of improving
the financial supply chain has been widely
recognized, approaches that companies and
software product vendors have taken so far
mainly focus on automating invoicing and
payment processes, for example, Electronic
Invoice Presentment and Payment [1]. A
holistic solution must go beyond automation.
The application of tried and tested tools from
the product supply chain world can bring
significant benefits to the company’s financial
supply chain.
The current environment of global
product and service flows creates even deeper
challenges and opportunities in the realm of
financial supply chain management. The trend
towards low cost country sourcing of products
and global outsourcing/off-shoring of services
means companies first focus on getting their basic
product/service supply chain right - in the face of
new infrastructural, informational and cultural
barriers. This leaves little time or attention to be
paid in optimizing supporting processes such
as financial supply chain management. These
trends also introduce inefficiencies in working
capital flows that can effectively increase the cost
of goods and services.
Global companies are recognizing
the need to better understand the challenges
and opportunities in increasing the efficiency
of their financial supply chains, especially in
the current environment of low cost country
28
sourcing and global outsourcing. We believe
there are useful approaches and methods from
the classical supply chain management realm
that can be innovatively directed towards
enabling improvements in the Financial
Supply Chain.
TRADE FINANCING IMPACT OF LOW
COST COUNTRY SOURCING AND GLOBAL
OUTSOURCING
The current shift towards low cost country
sourcing poses new challenges and opportunities
in the financial supply chain. The trend increases
the complexity and reduces the visibility of
events that contribute to the product-movement
and the associated funds movement across
the players in the supply chain. This situation
results in imperfect information on physical
and financial flows, causing the build up of
“cash buffers” much like inventory buffers in
the physical supply chain. Although companies
have invested in sophisticated tracking
mechanisms to pinpoint exactly where products
are in the supply chain, the same information is
typically not available or not leveraged in the
financial and cash management processes and
systems that govern optimal deployment of
working capital.
The cost of working capital in emerging
countries is typically higher than in western
markets. The level of sophistication of processes
and systems of suppliers is generally lower that
that of buyers in developed markets. While
buyers attempt to extend payment terms to
be more effective with their working capital,
suppliers must take on increased financing cost
to enable this, ultimately resulting in increased
cost of goods for buyers.
If we were to consider inventory
turnover ratio as a proxy for the efficiency of the
physical supply chain, the receivables turnover
ratio could be a parallel for the financial
supply chain efficiency. According to the CFO
magazine, between 1980 and 2000, the average
inventory turnover period in publicly listed US
corporations was reduced from 73 days to 48
days, a net saving of 25 days, while receivables
turnover decreased only by 11 days - from 68
days to 57 days [2].
At first glance, financial supply chain
optimization appears to be a far simpler problem
than its physical counterpart. Integrating
accounting information should be fairly simple
– in the most elementary sense, there is only
one SKU (dollars) and two possible locations
(buyers and seller’s bank accounts) for cash
in a supply chain transaction – making it
a more straightforward problem to tackle
– if trading partners, especially buyers are
motivated enough.
Therein lies the problem – buyers
are generally not motivated to make major
improvements in settlement efficiency. By
extending payment and credit terms, sellers
provide working capital finance to the buyers.
On the surface, this looks advantageous to the
buyer since it appears as though the buyer is
receiving free credit. The incentive for the buyer
to settle early is not transparent enough to create
an impact on the buyer’s strategies and plans to
improve working capital efficiency.
Imagine a situation where the cost of
capital for the buyer is 8% p.a. and that for the
seller is 15% p.a. This is not far from the truth for
current low cost country sourcing environment
[3]. Also imagine two scenarios, one in which the
buyer settles the seller’s invoice immediately (on
receipt) versus one where the buyer pays after 60
days. The second scenario requires the seller to
get working capital finance at a rate that is 7% p.a.
more than what the seller would pay to finance
the payable. The seller incurs an incremental
29
cost of approximately 0.7% of revenue as added
cost of finance. As the seller’s profit margins are
squeezed by globally liquid market for goods
and services, this cost will ultimately be passed
back to the buyer, sometimes with a profit mark-
up. This hypothetical scenario is vastly simplified
from reality, but conveys the essential challenge
facing companies today.
ANALYZING FINANCIAL SUPPLY CHAIN
OPPORTUNITIES
In order to get a clear picture of improvement
opportunities and available savings potential,
companies need to first translate the information
inherent in their product/service supply chains.
Traditionally, paper based processes have
been used to exchange business documents,
including invoice and payment information,
across business partners. These processes
provided little visibility into the actual status
of cash positions or requirements, limiting the
ability of companies to react to events causing
disruptions in the supply chain. Over the last
10 years, supply chain collaboration and event
management technologies have given rise to
higher level of visibility of schedules, status and
impacts both upstream and downstream due to
a disruptive event happening at any point in the
supply chain [4].
To achieve better visibility into
financial supply chain opportunities, a company
would need to fully understand the relative
significance of various working capital assets
and liabilities. The starting point is a visual
map that, that represents the assets and
liabilities forming part of the company’s
financial supply chain. Fundamentally, a
financial asset can be equated to inventory in
the physical supply chain. It is characterized
by its size (in dollars) and its duration
(term). The following steps outline a basic
 EXAMPLE PLOT OF WORKING CAPITAL ASSETS/LIABILITES

Duration time
LCC suppliers

Regional
Independent
Retail

Tier 1 National
Outsourced
Services
Mail Order

MRO Suppliers
MRO Suppliers

Size of Asset

Figure 1: Hypothetical Plot of Working Capital Assets/ Source: Infosys Research

Liabilities

approach to enable a company to classify and axis representing the settlement

characterize its working capital assets and cycletime
liabilities with a view to identify the areas that • For each cluster, draw a circle around
offer the most opportunities for improvement. the mid-point, with the radius of the circle
representing the variability of the
• Segment supply chain transactions into settlement cycle in terms of the standard
groups, based on characteristics such as deviation of the mean.
average values of receivables/payables The map [Fig. 1] enables the company to get
and actual settlement cycle times. better visibility and establish areas of
The segmentation could be performed opportunity. The idea is to segment the
according to customer or supplier opportunities on both the buy side and sell
groups, product families or geographies side. The most promising opportunity areas
• Map the groups into a scatter plot, with are those with high impact and high
the horizontal axis representing the variability as denoted by the larger circles
size of the asset/liability and the vertical in the top corners of the chart.

30
APPLYING CLASSICAL SUPPLY CHAIN
MANAGEMENT LESSONS TO FSCM
Most of supply chain improvement theory and
practice over the last twenty years has focused
on a few key themes – increasing responsiveness,
improving velocity, decreasing variability and
optimizing resource use. Approaches such as Six
Sigma, Lean, TQM and Theory of Constraints —
all focus on one or more of these related themes.
A tremendous amount of rigor and learning
has resulted from the systematic application of
these methods to the physical supply chain. It
is worthwhile considering the applicability of
such approaches to solve the challenges posed
by today’s global FSCM environments.
Approach 1 – Using Six Sigma techniques to
reduce variability
A significant amount of working
capital is locked up in un-reconciled invoices
and payments. Financial systems are geared
towards forecasting cash requirements on
the basis of expected payments over the
short-term time window. For instance, cash
requirements for servicing payables are
generally forecast over a 60-90 day period based
on historical patterns, as well as commitments
arising from purchase orders. However, if a
vendor invoice does not match the purchase
order or goods receipt, the invoice generally goes
into a “held” state, with some action required
either internally or from the vendor to enable
the company to post and pay the invoice. This
results in disruptions in the cash forecast - some
expected cash outflows do not happen, with
the resulting excess liquidity and unwarranted
cost of financing. Similar situations also happen
on the accounts receivable side, with invoice
exceptions occurring within customer systems
to cause cash shortfall and resultant need to find
short term cash at higher interest rates.
31
Optimizing accounts payable and
accounts receivable functions assumes critical
importance in the world of financial supply chain
management. Improvement initiatives aimed
at these functions have traditionally centered
around process automation and internal error
reduction. However, making real improvements
to the financial supply chain require a focus on
management and reduction of exceptions across
supply chain partners. Variability in settlement
cycles is the chief enemy here.
Transaction exceptions cause high
degree of unpredictability and attendant cost
increase in the financial supply chain. Exceptions
can run to as much as 5% to 10% of transaction
volume, lock up cash and increasing financing
costs. Six sigma approaches can be very
effectively leveraged to reduce exceptions and
reduce variability in the financial supply chain.
Case example:
By applying its six sigma methodology to
the analysis of invoice payment process and
implementation of a best-of-breed EIPP system,
General Electric realized the following benefits:
• 61% reduction in paper invoices
• Reduced invoice defect rates
• Average rework queue for reconciliations
reduced from 85,000 to 5,000
• 90% of allowable discounts captured,
worth over $100mn
• Improved ability to forecast cash flow
requirements.
Through this initiative, GE saved about
$1.8 bn or 12% of their Accounts Payable in one
year [5].
The approach starts by defining the
problem, i.e., inherent variability in process
resulting in exceptions. The variability is then
measured, through the use of a series of metrics
such as “sigma of process.” The variability can be
benchmarked, against other business partners,
locations or product types. For instance, it may
turn out that the rate of invoice exceptions
from supplier A are double that of supplier B,
although both of them supply roughly the same
commodity. An analytical phase follows, using
techniques such as root cause analysis and
correlation analysis. Business partners have to
work closely, leveraging cross-functional teams
from individual organizations to effectively
perform this type of analysis. Improvement
projects and action plans are then put in place
and executed, with close and collaborative
teamwork from both partner organizations
being necessary to make sure projects are well
planned and executed. Finally, the effectiveness
of the outcome is constantly measured through
use of control charts and other similar methods.
Approach 2 – Extending supply chain event
management to FSCM
The concept of supply chain event
management (SCEM) has been in vogue over
the last several years and relates to the ability of
organizations to respond intelligently to events
occurring in their supply chain that deviate
from planned outcomes. When extended to the
financial supply chain, this concept has powerful
parallels. For instance, delays in product receipts
and attendant delays in cash outflows that have
an impact on the cash forecast. Many supply
chain systems have the ability to incorporate
such delays into planning and execution cycles,
but this information is rarely actively considered
in financial activities such as cash forecasting.
In addition to information on physical supply
chain events, event management from a financial
perspective must also factor in information on
transaction exceptions, such as the example of
32
unmatched invoices mentioned previously. A
more accurate forecast of financing needs allows
the company to fine-tune cash requirements
enabling the most effective method of financing
outstanding cash requirements.
Some industries have adapted well
to this need of streamlining the material and
financial supply chains by defining new business
models. The vendor managed inventory
(VMI) model adopted widely in high-tech and
manufacturing industries is one such example.
In the traditional purchase order-based model,
the supplier sends an invoice as soon as the
material is shipped and the invoice creation-
to-payment cycle is long especially if the 3-way
match process (validating receipt of the material
and right price on the invoice), is not automated.
However, in the VMI process, the supplier is
paid by the buying organization when material
is consumed from a hub located in buyer’s
manufacturing facility, resulting in reducing the
payment cycle significantly.
With increased visibility to supply
chain events affecting the inventory, AR and
AP, organizations are able to manage working
capital effectively. One of the leading examples
of such a transformation is Dell Computers
which had a negative cash-to-cash cycle and
excellent visibility to forecast cash flow [6].
Cash to DSO (34d) + DII (6d)
Cash
Cycle =
(-12d) DPO (52d)
To effectively apply these techniques,
the basic requirement is to have a tight linkage
between the supply chain execution, AR/AP
processing and cash management functions of
the company. The information building blocks
do exist in today’s sophisticated generation ERP
environments such as SAP and Oracle. What
is needed, in most cases, is the organizational
processes necessary to synthesize this
information and translate it into action.
Approach 3: Applying optimization methods -
Dynamic Early Payment Discounts
Needless to say, there are constant innovations
and breakthrough ideas that are at work in
transforming the world of financial supply
chain. The concept of dynamic early payment
discounts is one of them, and has the potential to
shake up the world of working capital finance [7].
Conventionally, payment terms were negotiated
in a very static way. A supplier would offer a
Dynamic Early Payment Discounting method helps
both the buyers and the sellers optimize their working
capital and reduce cost of finance
standard payment term (e.g., Net-30) and an
alternative, discounted payment term (1%10
Net 30). In some cases, large and more powerful
buyers would enforce a set of standard payments
terms on their suppliers. These methods do not
recognize the relative differential value of funds
to the buyer and supplier. This is exacerbated in
the environment of low cost country sourcing
and global outsourcing, where the differences in
cost of financing are most pronounced.
In the dynamic discount environment,
buyers and sellers essentially allow a high degree
of flexibility to their payment terms, reflecting
33
the true cost of working capital and creating
a virtual marketplace for funds. Consider the
example, where a supplier has a standard
payment term of Net-30. The supplier, however,
is currently experiencing a cash crunch. The
supplier may then extend higher discounts for
early payments to its customers. A customer
that has excess cash may take up the discount
and settle early to get an effective cost saving.
This process would be too time consuming with
traditional paper based systems. However, given
a fully electronic environment where all invoices
and payments are accessible through a central
settlement network, sellers are able to offer
revised payment terms dynamically using a rules
based engine that is driven by the seller’s “cost
of working capital.” These electronic “offers”
are evaluated and automatically acted on by
similar engines that buyers run on the settlement
network. A negotiation process between supplier
and buyer on the date of payment of invoice and
associated discounts can also be enabled. This
type of dynamic discounting allows both buyer
and seller to optimize their working capital and
reduce costs of financing through enabling a free
market for working capital.
Although the approach sounds
somewhat theoretical, it is not far removed from
the capabilities of applications and tools available
in today’s environment. Companies may start
to take steps towards this state by automating
their invoicing and payment processes, enabling
the accounts receivable and account payable
departments to collaborate on-line and offering
more flexible discount options, besides the
standard ones.
To take advantage of such opportunities,
AR and AP departments need migrate to being
more innovative and less transaction driven.
Needless to say, trust between trading partners
is critical and this may mean that dynamic
discounting, at least initially, is restricted to
certain categories of customers of suppliers that
have the necessary strength of relationships in
place and offer the greatest opportunity for
optimization. It is also necessary for virtual
“marketplaces” for discount management
are in place. This void is rapidly being filled
by boutique software firms as well as global
financial institutions that are setting up and
propagating similar discounting platforms.
CONCLUSION
Translating product supply chain data into
information and knowledge improves the
predictability of cash flows and frees up unused
cash in the financial supply chain. Recognizing
and leveraging the relative value of cash to
different trading partners in the supply chain
and applying systematic techniques to improve
34
predictability of supply chain will result
in significant benefits for companies in the
increasingly global, networked environment
they operate in today.
REFERENCES
1. e-Billing.org (Resource for Electronic
Presentment & Payment) http://www.
ebilling.org/EIPP/eipp.html.
2. The Argument for Financial Chain
Management, CFO Magazine, December,
2000.
3. New Strategies for Financial Supply
Chain Optimization, Aberdeen Group,
November 2005.
4. “The Next 15” – Trends in Treasury &
Risk Management, Treasury & Risk
Magazine, September 2006.
5. Warren H Hausman, Financial
Flows and Supply Chain Efficiency,
Visa International White Paper 2005.
Available at http://corporatevisa.
com/md/dl/documents/downloads/
supplychain.pdf.
6. Dell’s Direct Hit, CFOAsia.com, May
1999, Available at http://www.cfoasia.
com/archives/9905-59.htm.
7. Dynamic Early Payment Discounts will
shake up Accounts Payable, Gartner
Research Brief, Jun 22, 2005.
 SETLabs Briefings
VOL 5 NO 3
Jul-Sept 2007

Tackling Supply Chain through

Grid Computing:
Opportunities and Challenges
By Sumit Kumar Bose and Shubhashis Sengupta

Lower costs and reduce analysis times for compute

intensive aplications in petrochem industry using
grid technologies

G lobal supply chains, especially for process

industries like petrochem industries, are
facing challenges of speed and scalability in
decision making and execution. From crude
exploration to crude procurement, production
planning, scheduling and distribution, a number
of operations characterize the petroleum supply
chain. Each of these operations, however, is
characterized by a great deal of uncertainty.
For example, crude exploration involves
millions of dollars of investment without being
absolutely sure about the crude availability
in a particular region. Furthermore, informal
practices prevalent in many of the production
plants result in inefficient operations
characterized by high inventories on some
occasions and shortages on other occasions.
Thus, there is a need for operational research
models that minimize the sum of procurement
cost, production cost, inventory holding cost,
deficit cost and distribution cost. However,
with increasing complexities and dynamically
changing business-operating conditions, there
is a need for faster decision making and supply
chain dynamics. The need of the hour is to have
globally linked and scalable IT infrastructure
that can capture, gather and process information
fast enough to make faster optimization
decisions. This means, the global petrochem
firms today have to run gigantic analytic and
simulation techniques that deal with millions
of variables and peta-bytes of data. Grid
computing is the answer to cater to such high
performance computing requirements. In this
paper, we highlight the analytic challenges in
today’s petrochem supply chain problems and
highlight an adaptive and agile Grid computing
infrastructure for the same.

35
CHALLENGES OF PETROCHEM INDUSTRY
Petrochemical and refining industries face
complex environmental challenges. To name
a few: reservoir uncertainty, global economic
growth rate uncertainty, geo-political risk, sub-
optimal production, inaccurate production
allocation, poor choice of crude are some of the
challenges that petrochemical companies have
to deal with on a regular basis. To illustrate
the nature of risks and uncertainties facing
these companies we elaborate on a few of these
below:
Reservoir uncertainty: Crude exploration
involves million of dollars of investment
without being absolutely certain about the crude
availability in a particular region. It is uncertain
whether profitable discoveries of crude can be
made at all.
Crude selection: The cost of crude forms a
significant percentage of the input cost for a
refinery. Additionally, the properties of the
crude obtained from different oil fields vary
significantly in their characteristics. Most
modern refineries, however, are quite flexible
and can process crude with widely varying
properties and are therefore faced with
multiple decisions regarding the purchase of
crude. The situation becomes complex when
multiple refineries are involved.
Crude loading and unloading: Crude purchased
from different locations is not of equal quantity.
Similarly, on the quality front, certain types of
crude are better suited for producing a product. It
is a challenge, therefore, to simultaneously decide
upon the proportion in which different crude
should be mixed in the charging tanks and the
portfolio of output products that will maximize
there finery’s profit.
36
Crude production: Efficient scheduling of
operation modes (also called run modes) in the
crude distillation units and the scheduling of
feeds and intermediates in different plants such
as the lubricant plant, the diesel manufacturing
plant etc., is critical for business’ profitability.
Inefficient schedules can lead to issues of large
inventories on some occasions and stock outs on
other occasions.
Trading and pricing: Energy trading plays
a critical role in a petrochemical company’s
integrated business. A host of geo-political
factors govern the crude price and crude output.
This in-turn impacts the profitability of the
petrochemical companies greatly. Petrochemical
companies thus need to manage the prices and
the volume risks.
Mitigating each of these challenges
requires complex operational research
models. Companies need to embrace complex
technologies and innovative models to mitigate
the different risks that arise in the various
stages of the decision making process of the
petrochemical supply chain process. Here we
identify the different decision making processes
within the petrochemical supply chain that are
amenable towards application of grid based
solutions.
GRID COMPUTING FUNDAMENTALS
Grid computing is a new and powerful
paradigm in distributed computing. It provides
a stack of software services that allows
interconnecting a large number of heterogeneous
computing systems distributed across wide
geographic locations. This in turn helps to
provide a uniform, virtualized single source
of computing to the user. It is an ensemble of
heterogeneous computing resources for
solving complex computation intensive
 1010

109

108 1Petaflop Grid Computing

107
Peak ThroughPut
(MFLOPS/sec)
106

105 1Teraflop
Cray T 90
4
10

Pentium
103
Pentium Pro

102 IBM
360/90 MAC

101 CDC IBM

6600 Aple II
Altair
1.0
8080
Inte
0.1 CDC 4004
6600

1960 1970 1980 1990 2000 2010 2020 2030

Year

Figure 1: Evolutionary Path of Computing Source: Adapted from books.nap.edu/html/

tech_21st/t2.htm

tasks. Two different forms of grid computing Evolutionary Path of GRIDs

can be recognized: Data Grid and Compute In the early 90s all high performance computing
Grid. Data grid is a manifestation of grid tasks were carried out on single node
technology that helps to achieve virtualization multi-processors (SMPs). The SMPs then gave way
of the data stored in multiple heterogeneous to clusters. Cluster typically consists of a number
databases stored across multiple locations. of SMPs usually located in a single location.
Additionally, data grids enable sharing The advent of high speed interconnects and
of computational load across different networking technologies have resulted in
machines. Compute grid provides the necessary collaborative computing and data sharing.
infrastructure for carrying out compute Collaboration and data sharing can take different
intensive tasks on different heterogeneous shapes. Figure 1 shows the evolutionary path of
machines simultaneously. computing.

37
 Crude Crude Crude Crude
Exploration Extraction Loading Transport
Figure 2: Petrochemical Supply Chain
Benefits of GRIDs
Adoption of GRIDs usually results in a number
of business benefits that include but are not
limited to:
• Real or ‘near’ real-time data integration
and delivery across heterogeneous data
sources
• Wider, faster end-user access to key
business data
• A single, comprehensive view of the
enterprise’s information assets
• Increased flexibility to change and scale
to environmental changes “on demand”
• Ability to analyze real-time business data
to accelerate decision making
• Greater re-use of data – EII infrastructure
built for one data integration project can
be re-used for other data integration
projects involving the same data sources
since data is only federated, not actually
moved
• Extension of the range of data warehouse
queryingfor better decisions and BI __
combine data warehouses query on
historic data with query on
real-time business data to make better
decisions.
38
Crude Refinery Distribution Trading &
Unloading Production & Shipment Marketics
Refinery Operations
Source: Sumit Bose, PhD Dissertation, IIM-Calcutta
PETROCHEMICAL SUPPLY CHAIN
The petrochemical supply chain can broadly
be categorized into five different segments:
exploration, extraction, refining, marketing
and distribution. Figure 2 shows a typical
petrochemical supply chain. Most companies
operate in all the five segments. The refineries
are owned by integrated oil companies having a
variety of interests, from exploration and
production through refining and marketing to
retail sales.
Crude exploration: This phase involves
locating possible crude reserves. This is a phase
where uncertainty involved for an integrated
petrochemical company is the maximum. Crude
exploration requires huge amounts of investment
without being absolutely certain about the
amount of crude that will be available in a
particular well. It is uncertain whether profitable
discoveries can be made by the investing
company at all. For these reasons, companies
need to accurately identify the profitable oil
fields before they undertake the actual drilling
operation. For this purpose, companies employ
geologists and geophysicists to analyze the
different seismic data collected from remote
sensing satellites.
 Activities such as seismic data pre-
conditioning, pre-stack depth and time migration,
velocity modeling and full wave processing make
use of this seismic data. Furthermore, sub-surface
attributes and rock properties of reservoirs need
to be determined for reservoir modeling and 3D
visualization. Each of these analysis activities
however is compute-intensive and holds great
promise for applying grid concepts for reducing
the analysis time and lowering the total cost of
carrying out the operations.
Crude extraction: It is pertinent that the
production of oil and gas is carried out at
maximum efficiency throughput on a 24x7 basis
and it is important to minimize the downtime
that may result due to repair and maintenance.
This requires surveillance of the oil well and
analyzing the performance of the drilling
operation on a real time basis so as to identify
the equipments that are likely to fail even before
the actual failure of equipment. Such real-
time reservoir monitoring (RTRM) to optimize
production requires management and analyses
of high volumes of data that flow in from the
automatic sensors, integration of different
kinds of data and visualization of the data in
conjunction with the 3D reservoir models. This
requires processing of continuous queries over
continuous data streams and is highly compute
intensive task that requires high performance
computing.
Refining: It concerns the scheduling of
operation modes (also called run modes) in
crude distillation units and the scheduling of
feeds and intermediates in different plants such
as the lubricant plant, the diesel manufacturing
plant, etc. Typically a crude distillation unit can
be operated in several modes and a changeover
from one mode to another has an associated
39
cost. Scheduling decisions involve deciding the
run modes in each unit in every time period that
would optimize a given criteria.
Optimization involves trade-offs
among cost of changeover, cost of inventory
and tardiness (or upliftment failure) penalties.
Associated with the refining process are the
two problems of refinery blending and refinery
pooling. Refinery blending involves producing
a large number of products by blending a small
set of basic grades in varying degrees. For
example in gasoline blending, feed with widely
varying attributes have to be mixed in different
proportions to yield products with different
attributes. Decision making here involves
selecting a set of basic grades (feed), deciding
the quantity that need to be produced for each of
the basic grades and the manner/proportion in
which the grades should be blended to meet the
final demand.
Refinery pooling problem is a
generalization of the refinery blending problem
and is used to model the mixing of different feed
in intermediate tanks. The problem of refinery
pooling is stated as follows: “Given the availability
of number of feeds, what quantities should be
mixed in intermediate pools to meet the demands
of various final blends whose attribute qualities
must meet the known requirements?”[1]. While
the refinery blending problem can be modeled
using linear programming approach, modeling
the refinery pooling problem requires nonlinear
programming skills. Optimization of the refinery
pooling problem has been the subject of research
for long. Large companies with multiple plants,
processing different kinds of crude and having
a diversified portfolio are faced with such
difficult optimization problems. Such problems
are extremely difficult to solve and are often
beyond the capacity of the present day solvers.
The solvers are capable of solving only limited
versions of the problems. This requires super-
linear speedup of parallel search algorithms
on super-nodes. Grid computing for parallel
algorithms provides a viable solution for solving
such hard problems.
Distribution: Distribution involves preparing
a shipment plan. Shipment plan generation
requires trade-off between cost of shipments
and cost of inventory at the stocking points.
Operational research models for the shipment
planning problem are hard to solve and this
Petrochem companies manage price and volume risks
by resorting to advanced analytics to forecast prices and
model volatility
requires use of parallel search algorithms on grid
for locating ‘optimal’ or ‘near-optimal’ solution
for large scale problems in real time.
Trading and marketics: Energy trading plays
a critical role in a petrochemical company’s
integrated business. The companies resort to
advanced analytics for forecasting prices and
modeling volatility. By doing so, petrochemical
companies try to manage the prices and the
volume risks. Companies undertake scenario
analysis to determine the different possible
scenarios that may arise and determine
risk mitigation strategies for the worst case
scenarios.
Scenario analysis is extremely useful
for analyzing performance of companies under
subjective views. It determines the outcome of a
40
strategy under one or several possible scenarios.
Most of the petrochemical companies therefore
undertake scenario analysis to understand
better on how the different geo-political
factors are going to impact their business.
Companies usually combine scenario analysis
with monte-carlo simulation to determine the
probability distribution of outcomes when all
possible market scenarios are taken into account.
Simulation however, is a complex compute
intensive task and is amenable for calculation
on a grid.
As already stated a number of other
decision making problems that arise in the
refinery context involve complex operation
research models. For example, let us revisit the
crude selection problem. As earlier mentioned,
the cost of crude forms a significant percentage
of the input cost for a refinery. Additionally, the
properties of the crude obtained from different
oil fields vary significantly in their characteristics.
Most modern refineries, however, are quite
flexible and can process crude with widely
varying properties. As a result, such refineries
are faced with multiple decisions regarding the
purchase of crude. For example, refineries enter
into long term contracts with the suppliers of
crude and buy from the future markets apart
from purchasing from the spot markets. As the
cost of crude forms the majority of the input
cost finding out the optimal mix of crude is
important. The situation becomes complex when
multiple refineries are involved. Moreover,
crude purchased from different locations is not
equal. Further, certain types of crude are better
suited for producing a product. It is a challenge,
therefore, to simultaneously decide upon the
proportion in which different crude should be
mixed in the charging tanks and the portfolio of
output products that will maximize the refinery’s
profit. Each of these operational models are
not amenable towards solution in real time by
present day solvers and hence require high
performance computing techniques that can
make use of parallel search algorithms.
ROLE OF GRID TECHNOLOGIES IN
PETROCHEMICAL SUPPLY CHAIN
In the crude extraction phase monitoring
applications need to monitor large volumes
of data flowing in from different sensors.
The applications need to analyze this large
data set in real time. Since data from sensors
come in continuously, queries that need to be
executed are often long running as against the
one time queries executed against persistent
databases. Traditional databases systems are
often ill equipped to handle such scenarios
involving continuous flow of data streams. To
take care of such situations a new paradigm
involving continuous queries has evolved.
The new paradigm allows continuous queries
to be executed over continuously changing
data streams. This is distinctly different from
querying persistently stored data sets. Aspects
that need to be considered for continuous queries
over continuous data streams are distinctly
different from the traditional query optimization
techniques that are done for persistent databases.
The paradigm of continuous queries allows the
managers to register specifications in which they
41
are interested over streaming data sources. The
continuous query filters out the irrelevant data
to provide the managers with the results that the
manager is interested in. Since data sources are
inherently distributed, multiple instances of a
query can be executed in parallel to process the
incoming data faster.
Many of the resource allocation
problems that arise in the refining process are
discrete optimization problems and can be
modeled using 0-1 binary variables. Similarly,
the different decision making problems that
arise in the distribution process can be modeled
using mixed integer linear programming models.
However, these models are often computationally
intractable and are hard to solve using sequential
solvers. The combinatorial search methods used
by most solvers are based on branch and bound,
branch and cut and branch and price algorithms.
These algorithms suffer from the problem of
combinatorial explosion i.e., the search space
that needs to be explored grows exponentially
with problem size and the algorithms therefore
require a prohibitive processing time. In
other words, the size of most of the problems
makes them intractable and beyond the scope
of the present day solvers that are based on
sequential processing. In these scenarios a
parallel computer can speed up computation
significantly by processing different computation
tasks simultaneously. Parallel computers help in
reducing the execution time and provide new
avenues to solve hard combinatorial problems
that are difficult to solve by sequential computers.
Solving such combinatorial optimization
problems using parallel computers (clusters)
requires designing parallel search algorithms.
Good parallel search algorithms can effectively
use the parallel computers and clusters for
solving such hard combinatorial optimization
problems. Parallel search algorithms reduce
execution times and offer tremendous speed-up involved in reservoir modeling and that in risk
over serial algorithms. Problem instances that hedging.
were regarded as computationally intractable
can now be solved using parallel algorithms CONCLUSION
running over small workstation clusters. Grid technologies offer promising hope
Simulation modeling is routinely for bring down cost and analysis time for
carried out as an adjunct to mathematical different compute intensive applications across
modeling for which closed form analytical the petrochemical supply chain. Possible
solutions are difficult to obtain. For example, applications of grids include: carrying out
reservoir modeling requires simulation the simulation and processing tasks in crude
experiments to be carried out for determining exploration, integrating different types of data
reservoir properties. Furthermore numerical and framing continuous adaptive queries for
methods like monte-carlo simulation are querying continuous data streams flowing in
routinely used for pricing and hedging optional from different sensors during crude production,
products in trading and marketics. Simulation using parallel search algorithms on parallel
however is a complex computation intensive machines for determining resource allocations
task. The numerical resolution of a system of and simulating different scenarios of the monte-
partial differential equations using finite carlo approach for carrying out risk analysis and
difference and finite element methods require energy trading.
the development of high-performance numerical
simulation tools that can efficiently exploit REFERENCES
computational grids. Monte-carlo simulation 1. Sumit Kumar Bose, Scheduling in
techniques are inherently parallel and the Continuous Process Industry: Models
application requiring monte-carlo simulation can and Solution Procedures, PhD
be effectively executed on grid. The computational Dissertation, IIM Calcutta, 2006.
resources of the computational grid can be 2. Anirban Chakrabarti and Raghavendra D.
used to conduct simulation experiments for Rao, Process-specific GRID technology
reservoir modeling and for determining prices solutions for the Petroleum Exploration
of products in the futures and forwards markets. and Production (E&P) business, Technical
Simulation allows one to better understand the Report, Infosys Technologies Limited,
inter-relationships between different parameters 2005.

42
 SETLabs Briefings
VOL 5 NO 3
Jul-Sept 2007
Learning from Supply Chain-
Mistakes and Experience
By Narayanan Sampath
Identifying and eliminating mistakes in
supply chains could lead to unlocking of
substantial value
S upply chains are typically associated with
a constant tussle of bringing together the
mismatch between demand and supply. Formed
out of a large number of visible and invisible
links, supply chains connect within and outside
organizations, between people of different
background, skills, and knowledge levels.
Typically, each component of the chain may be
driven by different constraints and priorities.
The diversity of deliverables at each link
along the chain makes it highly vulnerable to
mistakes. The mistakes are further exacerbated
by the urgency with which activities are carried
out to fulfill ever changing demands. It may
not be an overstatement to state that even
in the best run supply chains, a number of
mistakes continue to occur, get repeated and go
unreported.
Conventional approaches to supply
chain management relate to improving
43
efficiencies of various events and processes
across the supply chain. These may be in the form
of improved processes, enhanced collaboration
between partners, use of technology, etc. In
practice, however, increasing efficiencies in an
operational supply chain requires larger efforts,
time and perhaps investments. While this is an
essential continuous improvement exercise, it
must be realized that such improvements cannot
be achieved in the short term. These require
medium to long term time horizons to bring
about visible results.
On the other hand, identifying and
eliminating mistakes in an existing supply chain
activity or process could result in simple and
sometimes dramatic improvements in a very
short span of time. This could be achieved with
much less expense of effort and may not always
call for a change in process or investment in
technology.
The focus would be to perform supply chain
activities minus the mistakes that have gone
unnoticed hitherto or been noticed but given a
go by in the pressure of the situation.
The succeeding sections of this paper
bring focus on typical mistakes that occur in
certain specific areas of supply chain execution.
Examples from the field highlight as to how these
mistakes impact the overall efficiency of the
supply chain. These are based on the author’s
personal experiences in running a global supply
chain in a pharmaceutical company. Strategies to
avoid if not eliminate these mistakes have been
suggested.
The mistakes are covered under the following
categories:
• Master Data at the core of a supply chain
• Low value items require greater
attention
• Delayed disposal can cost dearly
• Mistakes in Packaging and
Transportation
• Ignorance of rules, regulations and
procedures is expensive
• Mistakes in Letters of credit.
MASTER DATA AT THE CORE OF A
SUPPLY CHAIN
The Mistake and its Impact
Master data constitutes one of the most
important but grossly underestimated aspects
that directly affect supply chain efficiency.
Incomplete or inaccurate master data is the
root cause of many a problem downstream in
the supply chain. As business and product
lines in a company grow, huge volumes of
master data get generated. These may pertain
to customers, vendors, raw materials, finished
products, spares and accessories, accounting
44
information etc. Every change however
minor necessitates meticulous review and
incorporation of changes to the master data. The
seriousness of this work is underestimated in
many companies.
Exhibit 1 illustrates the range of SAP
master data to be maintained for different types
of materials. There are 23 different views to
be filled in and each view comprises of several
fields that affect the manner in which the
material is planned, ordered, stored, issued
and accounted. The ownership of each view
rests with different departments that need to
coordinate among each other to ensure that the
overall master is maintained completely and
accurately.
In practice however, master data
management is a part time activity done
by a set of people who work in different
functions. Absence of coordination, attention to
detail and review mechanisms lead to serious
downstream mistakes in the supply chain.
Procurement and production personnel are
often seen groping with problems related to
incorrect master data or are chasing different
groups to create a new data to enable them
to proceed with their activities. For example,
in the pharmaceutical industry, a minor error
in the number code of a raw material in a
product or assigning an incorrect specification
to a material can result in the procurement of
the wrong raw material and manufacture of
products that do not conform to their approved
specifications. Consequently, this can lead to
rejections, product recalls and in serious cases
even legal claims. In the retail industry, mistakes
in addresses of people, companies or vendors
have led to missed delivery of shipments. On
the HR front, incorrect master data can lead to
major errors such as crediting incorrect salaries
to employees.
 Semi-
Raw finished Finished
Sl No Department SAP View Material material product

1 Basic Data • • •
2 Design Basic Data 1 • • •
3 Classification • • •
4 General Plant Data Storage 1 • • •
5 General Plant Data Storage 2 • • •
6 Purchasing •
7 Purchase Foreign Trade Import •
8 Purchase Order Text •
9 MRP1 • • •
10 Purchase/ MRP2 • • •
11 Planning MRP3 • • •
12 MRP4 • • •
13 Planning Work Scheduling • •
14 Quality Quality Management • • •
15 Accounting 1 • • •
16 Finance/ Accounting 2 • • •
17 Accounts Costing1 • • •
18 Costing2 • • •
19 Sales Org. Data 1 •
20 Sales Org. Data 2 •
21 Marketing Sales General Data •
22 Foreign Trade Export Data •
23 Sales and Distribution Text •

Exhibit 1: SAP Material Master Source: Infosys Research

Suggested Strategy regular reviews, manage change control and

Master data needs to be seen as a core area
that requires excellence built around it. Senior
managers need to be well informed of the
importance of master data and the impact it can
have along the supply chain. It pays to assign a full
time cross functional team that would govern all
aspects of master data – setting policies, conduct
create or modify data in a time bound manner.
LOW VALUE ITEMS DESERVE GREATER
ATTENTION
The Mistake and its Impact
Often, lack of attention to low value items can
derail multi-million dollar shipments. While the

45
A and B class items always demand and get the
required care and attention, C class items very
crucial to the delivery of a product get neglected
and their requirement is realized very late.
Major shipments are affected due to the
non availability of C class items. There have been
instances where printed labels for a sterile injection
consignment reached late or were printed wrongly
and had to be redone. The total value of the labels
was of the order of US$ 400 but without them
being available on time, the production could not
be commenced leading to delayed shipment of
the high value consignment.
In another instance, unusable wooden
pallets discovered at the time of packing led to
the shipment getting delayed. The value of the
pallets was USD 300 and the consignment USD
350,000.
Suggested Strategy
Both instances stated above reflect inadequacies
in the supplier network, absence of planning and
preparation for a shipment and above all the
inability to attend to minor details. It is essential
to identify, monitor and ensure the availability
of low value items that have a potentially large
impact on high value shipments.
DELAYED DISPOSAL CAN COST DEARLY
The Mistake and its Impact
Even the most well balanced supply chains
end up having non-moving inventory. It takes
supply chain managers considerable time to
dispose off non-value adding inventory. They
are either constrained by procedures that prevent
them from taking a disposal decision or they delay
the decision in anticipation of an oncoming order
or for the fear of accepting a wrong decision and
being ticked off. In any case, such delays cost
money. In food or pharmaceutical industries
where materials have shelf life and expiry
46
periods, these delayed decisions are associated
with substantial costs. There have been instances
where managers have sat on disposal decisions
of non moving inventory for months and were
eventually forced to dispose them at zero value.
This loss is without accounting for the inventory
carrying cost.
Suggested Strategy
It is important to recognize non moving inventory
and take timely disposal decisions. This is a
painful decision to take for the supply chain
manager but experience reveals that realizing
partial value of an unused material is better than
getting nothing at all.
MISTAKES IN PACKAGING AND
TRANSPORTATION
The Mistakes and their Impact
Printing of packaging materials, the physical
activity of packaging the finished product,
loading them onto containers and ensuring safe
transit constitute some of the routinely carried
out activities in supply chain. However, these
activities are prone to some serious lapses as can
be seen from the following examples.
Printing Mistakes
Global trade requires organizations to ship
goods to other parts of the world with packaging
printed in the local language of the country being
shipped to. This requires immense amount of
care as the vendor supplying printed materials
has to equip its staff to be able to read and certify
the appropriateness and correctness of the
printed material.
In one instance, the printed labels of
a consignment to Japan got interchanged with
those belonging to another product. With the
stores and production line unable to comprehend
the Japanese characters, the label was issued
for production. The mistake was noticed by a
quality inspector at the very last stage before
shipping and the entire consignment was
rejected. The possibility of the material being
shipped out with incorrect description on the
labels could have led to disastrous consequences
but for the alertness of the quality inspector. The
root cause of the mishap was traced back to the
supplier of the printed packaging materials in
whose facility, two production lines got mixed
up and the absence of expertise and familiarity
Packaging has always been a problem area in the
supply chain continuum, largely because of oversight
due to rush in fulfilling orders
with Japanese characters led to it being passed
undetected.
Packaging Inadequacies
Despite advances in packaging technology,
packaging often gets lesser importance in practice
as people rush to fulfill orders. A consignment
of high value sterile injectable raw material for
a pharmaceutical company shipped from China
had substandard external packing material that
was damaged in transit. The entire consignment
was rejected by Quality as no chances are
taken with materials that are injected into the
bloodstream of a patient.
In another instance, poorly packed
high value capital equipment was found
seriously damaged due to ingress of water into
the main machine and its key components.
While the company did recover part of the
47
losses through insurance cover, the process
of photographing the damaged components in
the presence of insurance inspector, submitting
the claim, answering queries between the
supplier, transporter and the insurance company
and expediting consumed considerable time
and effort. Besides this, the project was set off
by over three months as the commissioning of
the equipment was delayed. The downstream
loss due to delayed commissioning of the
facility ran to several thousand dollars. This
entire loss could have well been avoided if
adequate care had been taken in packing the
equipment.
Transportation Mistakes
Damage to products due to poor packaging
of shipping containers is not an uncommon
problem. In rushing to meet shipment deadlines,
less attention is paid while stacking pallets
or shippers. The resultant damages cause
rejections by customers and consequential loss
of revenue and customer dissatisfaction. In one
such incident several containers containing
expensive drugs shipped to Australia were
stacked with large gaps in between without
intermediate packing. Movements during
transit caused physical damage to the packages
leading to a large part of the consignment
getting rejected.
Suggested Strategy
It is to be expected that in global trade, products
will transit thousands of miles across geographies
through different modes of transport. They would
be subjected to rough handling in ports and are
also likely to be exposed to elements. Hence, it
is essential that due care is taken in packing and
stowing them. As opposed to treating packaging
as a routine activity carried out at the end of a
production line, companies need to focus on
achieving excellence in all aspects of packaging
and delivery. Additional investment in superior
quality packaging material, improved methods
The tedium associated with understanding EXIM
management can create a huge potential bottleneck in the
downstream supply chain
and a higher control on packaging quality and
transportation methods may prima facie seem to
increase the costs but are investments that would
pay back very soon.
IGNORANCE OF RULES, REGULATIONS &
PROCEDURES IS EXPENSIVE
The Mistakes and their Impact
Institutions that train people in foreign trade
invariably focus on the macro aspects of logistics.
However, imports and exports management
is hardly as glamorous as it is perceived by
many. It is a job that involves a great degree of
monotony – being aware of rules and regulations
that vary from country to country, subject to
frequent changes as government policies and
relationships change and filling in a plethora of
48
forms accurately. A small mistake can prompt a
lower level functionary to disallow a high value
shipment and result in delays and demurrages.
Clearance of goods in ports for imports
or exports especially in developing countries
is an arduous task – prone to bureaucratic and
procedural delays, multiple hand-offs between
customs, airport/ port authorities, airlines/
shipping companies, clearing agents, etc. While
a large part of these procedures are handled
fairly competently by third party logistics
service providers, their work is also dependent
on the accuracy and completeness of documents
furnished by the exporter and importer who
operate across different continents. The examples
cited below indicate how ignorance of governing
procedures can create downstream snafus in
supply chain.
Incomplete Documentation
Global shipments need to have accurate and
fool proof documentation accompanying them
so that they pass through various procedures
and checks at different check points and reach
their destination in the shortest possible time.
The documents may range from invoices,
material description and specifications,
certificates of origin, bill of lading/ airway bill,
special permits for controlled substances etc.
While third party logistics providers do help
in ensuring these requirements, the importing
or exporting company needs to be fully aware
and equipped to provide the complete fool proof
documentation set. Often, missing or defective
documents cause serious delays in the movement
of the consignment. The primary reasons for this
problem are:
(a) Ignorance of rules and regulations
(b) Inadequate attention to detail leading
to incomplete documentation
(c) Mismatch between related documents
and typographical errors. For instance
the supplier’s invoice may carry a
material description which is different
from that stated in the purchase order.
Importing Controlled Substances
Companies that import or export controlled
substances need to be aware of the procedures
involved in doing so. Some of these procedures
are fairly long drawn, involve multiple agencies
and may involve substantial lead times.
Import of materials containing narcotic
constituents (typically used in pain killers) is
a case in point. The importing country has to
have the necessary quota assigned to it by the
International Narcotics Bureau that controls
the movement of narcotics. The destination
country’s governing body then allocates the
licensed quantity to the manufacturing company
through its own internal regulations. Once the
necessary clearances and documentation has
been obtained, the movement of the material
is strictly monitored till it is received at its
destination. All these procedures may take 90 to
120 days. Unless the supply chain manager plans
every single activity, the order execution may
get severely delayed. The transaction involves
government departments in the countries of
the exporter and importer, the diplomatic office
49
in the importing country, the regulatory and
drug control authorities and the airport/ port
authorities.
Materials such as Glyceryl Trinitrate
used in the prevention of heart attacks are
covered by rules governing use of explosive
materials and hence require special approvals
and precautions during transportation. These
therefore require longer lead times.
Similarly exporting drugs to some
countries requires prior approvals from drug
control authorities. Documents need to contain
details of drug being exported, its specifications,
precise quantities being exported and a host of
other details that need to be filled in. Shipments
will have to be planned keeping the lead times
required for approvals. Else, consignments ready
for dispatch get delayed and incur demurrages.
Incomplete Insurance Cover
Lack of attention during the initiation of insurance
cover is one of the commonly seen mistakes.
Although the people concerned do this as a part
of their routine function, the lack of attention
to detail leads to delays and costs money.
Consignments are reached to their eventual
destination through multiple modes of transport.
The transportation to the nearest seaport or
airport is almost always automatically covered
by insurance. People however miss out the “last
mile” – transportation from the port to the final
destination by road although this constitutes the
smallest component of the overall journey of the
consignment and at a minor incremental cost.
The result of this mistake is that the company
ends up taking an additional insurance cover
for the road transport at a higher price and the
consequent loss of time. Sometimes, when this
mistake goes unnoticed and the consignment
gets damaged in road transit, the losses are
substantial.
 Typically a “warehouse to warehouse”
insurance cover is the answer to the problem
but lack of knowledge and inadequate
communication on insurance requirements lead
to repeated mistakes. Personnel engaged in
sales, purchase and accounts need to be specially
trained in all aspects of practical insurance.
Insurance policies need to go through a strict
checklist-based quality check before being
initiated.
Suggested Strategy
The above examples highlight the importance
of the need to understand the nuances of every
activity related to exports, imports, insurance
and such formalities. While some of the examples
pertain to the pharmaceutical industry, each
industry has its own special requirements.
Therefore, the process associated with every
material or product may have subtle variations
and supply chain managers need to understand,
document the procedure and have checklist
aided control systems to prevent mistakes.
The supplier and customer networks also
need to be sensitized to the specifics of policies
and procedures so that all concerned parties
understand and fulfill their part of the obligation.
Needless to add, rigorous training of personnel
on the specific nuances of their supply chain
would go a long way in preventing expensive
mistakes.
MISTAKES IN LETTERS OF CREDIT
The Mistakes and their Impact
Letters of Credit (LC) constitute one of the most
accepted forms of financial transactions in global
trade and also one of the areas where a number
of mistakes are invariably made repeatedly.
If one were to consider letters of credit to form
their own supply chains comprising of exchange
of documentation and information across
50
parties within and outside the organization, one
can notice mistakes occuring in almost every
link in the chain.
Mistakes occur when a purchase
order made by the procurement department
with the payment terms specifying LC terms
are transferred to the finance department for
processing with the bank. Typographical errors
are quite common in a host of details that
need to be filled in like description of materials,
consignee name and address, bankers’ address,
swift code etc. Mistakes in stating the transaction
currency, units of measurement for the goods
being supplied (kg, lbs, litres) also are made
as the finance person is blissfully unaware
of the requirements. Then there are errors at the
bank end on either side further compounded
by transmission delays.
For instance, LC transmission to a
bank in Reykjavik (for procurement of fish
oil or cod liver oil from a supplier in Iceland)
is routed through a bank in London or
New York. Invariably the LC gets delayed at
this transit point due to mistakes, incorrect
communication and absence of follow up.
The hold up comes to light only when the
supplier expresses inability to supply due
to non-receipt of the LC and this starts a
frenetic search for the missing LC. By then the
shipment is delayed.
Delayed LCs affect the buy side as well
as the sell side of supply chains. On the buy side,
mistakes and delays in LCs result in delayed
shipments and affect production and delivery
schedules. On the sell side, delayed LCs affect
the financial supply chain and create fund flow
problems. Besides, correcting mistakes in LCs is
an expensive affair as banks charge significant
sums of money to incorporate changes. For
many companies this constitutes a tidy sum of
money which is an avoidable waste.
Suggested Strategy reference to the company’s specific
LC based business transactions call for needs would pay back very soon.
specialized knowledge across different functions 4. Document mistakes and build training
in the organization. Checklists and systematic programs that highlight the impact
handoffs are necessary to prevent trivial errors. of mistakes; train and retrain every
The people involved are also to be rigorously group involved to understand the
trained in the nuances of dealing with LCs. impact.
5. Build checklists for every single
RECOMMENDATIONS activity. These lists need to be
The mistakes in supply chain presented through understood and implemented
the author’s personal experiences are neither rigorously by internal functions,
those that have not been done before nor those suppliers and partners.
that are unknown to supply chain practitioners. 6. Effective supply chain planning is not
However, it is a fact that mistakes of this nature only about using IT tools and ERP
and more continue to occur in organizations in systems to carryout an integrated
one form or another along the chain at different planning and production scheduling. It
links. The vastness, variety, dependencies should also be able to proactively
and complexity of activities and participating anticipate various factors that are
constituents in a company’s supply chain not captured by any system but result
increases the probability of mistakes being in expensive mistakes and consequent
committed at every step. Based on experiences, delays. These should be factored
the recommendations of the author to supply into the planning process to ensure
chain managers are: higher delivery predictability.
7. Set up a group that specializes in foreign
1. Accept occurrence of elementary trade compliance. This group’s primary
mistakes in supply chain to be task would be to create a knowledge
an inevitable fact. Proactively focus on repository specific to the company,
improvements through mistake understand the specific compliance
elimination. requirements, build documentation
2. Set up a task force only to identify and processes around them and train
potential failure modes and the execution teams.
continuously work towards correction 8. Regularly interact with customers,
and elimination. suppliers and partners who contribute
3. Managing global supply chains requires to the supply chain. Make them
a comprehensive knowledge of rules, understand the impact of the mistakes
regulations, procedures and made at their end and initiate steps to
documentation. In practice however, have them corrected.
a large part of this knowledge is acquired 9. Quantify the cost of mistakes and
only through experience, particularly regularly publicize the savings
those gained from committing mistakes. achieved through avoidance of
Investing in rigorous training with mistakes. Reward groups and

51
 individuals that ensured mistake-free 3. Hau L. Lee, The Triple-A Supply Chain,
transactions. HBR OnPoint Enhanced Edition, May
2007.
CONCLUSION 4. New Strategies for Global Trade
Managers running global supply chain virtually Management – How enterprises are
live on an edge. One can never really predict mastering cross-border supply chains
when and where mistakes would be made and by synchronizing logistics, compliance
derail a well architected plan. Be that as it may, a and finance, Aberdeen Group, March
special eye towards spotting potential mistakes 2005.
and nipping them off before they can cause large 5. Beth Enslow, Best Practices in
scale damage is one strategy that will eventually International Logistics–How Top
pay rich dividends. companies use technologies and logistics
partners to improve performance,
REFERENCES Aberdeen Group, January 2006.
1. Martin Christopher, Logistics and Supply 6. Quantifying the impact of supply
Chain Management: Creating Value- chain glitches on shareholder value
Adding Networks, Financial Times – The significance of adaptive supply
Prentice Hall, 2005. chain networks, SAP White Paper, 2003.
2. B S Sahay, Supply Chain Management: For 7. Sumantra Sengupta, The Top Ten Supply
Global Competitiveness, Macmillan Chain Mistakes, Supply Chain
India, 2004. Management Review, January 2004.

52
 SETLabs Briefings
VOL 5 NO 3
Jul-Sept 2007

Global Trends in
Supply Chain Planning in
Semiconductor Industry
By Arnab Banerjee

The global semiconductor industry needs

a very high level of planning and
collaboration between multiple stakeholders to
ensure a smooth supply chain

O ver the years, a new world has emerged
__ an electronic world of computers, the
internet, wireless communications, satellite
navigation, high-tech machines, newer
entertainment styles and hundreds of other
pieces of wizardry. For the most part, this world
has little contact with the traditional physical
realm of earth, water, fire and air; indeed, one
of the great appeals of the cyber-world is its
ability to transcend time, distance, and material
limitations with invisible digital data.
The point of contact between the
electronic and physical world is the factory where
semiconductors are made – the wafer fabrication
facility or better known as “Wafer Fab” and the
human minds, which turn the very basic natural
material into a cyber world. In essence, wafer fab
takes very pure silicon (the basic element in sand)
modifies it with phenomenal dexterity, precision
and knowledge, and produces integrated
circuits __ the physical devices that power the
cyber-world. Doing this requires leading edge
capabilities in optics, chemistry, mechanics,
measurement and many other fields. It demands
huge amounts of capital __ both financial and
intellectual. The complex process constantly
redefines its own limitations, and somehow
manages to move ahead at a blinding rate.
Supply chain planning: Semi-conductor
manufacturing industry is ensnared with
complicated manufacturing processes. As
semiconductor business is rapidly changing
with variable demand, obsolescence, capacity
concerns, global manufacturing setup, contract
manufacturing and multi-tier suppliers playing
an important role in the over all supply chain,
the planning process has become very dynamic,

53
complicated and highly sophisticated. This paper As a definition of a classical supply
attempts to analyze the global trends in supply chain, three aspects of the overall supply
chain planning and that of factors influencing chain are examined __ operational supply
the global trends in planning such as business chain, inbound supply chain, and outbound
models, and product kinds. Semiconductor supply chain.
industry is characterized by extensive networks
of supply chains with both in-bound and out- Operational supply chain: This is the most
bound partners. Generic supply chain software important part of supply chain modeling
seldom satisfies all the demands of a complex especially for a wafer fab. It becomes all the
network of multiple partners and raises various more critical because of complex and resource-
issues concerning reliability. intensive manufacturing processes and various
The paper therefore also dwells on other factors such as high lead time, fluctuating
the mechanics of supply chain planning in two demand, complicated BOM and routing
scenarios — collaborative planning suite and structures, variable yield structure, presence of
non-collaborative setup. contract manufacturing, short life cycle of the
finished goods, and so on.
SC IN SEMICONDUCTOR INDUSTRY As operational supply chain plays
Semiconductors are a wide variety of products a major role, its planning is integrated with
and are used in varied products such as wireless transactional process packages, which are very
handsets, display and imaging devices, often the standard ERP packages in a majority of
information infrastructure and information well-established semiconductor companies.
access devices such as mobile and cordless
phones, flat-panel and conventional CRT Inbound/ outbound supply chain: Logistics
displays and monitors, high-speed local area plays an extremely important role in the industry
networks, TV set-top boxes, DVD and CD as it caters to a global customer base and adopts
players, internet access devices, and a host of assorted strategies such as global manufacturing,
other electronic systems. As result of its wide contract manufacturing, shared manufacturing,
ranging product portfolio and its labor- and and multiple sourcing/suppliers.
technology-intensive processes, the industry The Inbound/ Outbound supply
is globalized to exploit the comparative chain is of crucial importance to both fab
advantages. Typically, the wafer fabs and chip and fab-less semiconductor companies.
assembly are at different locations and the chain Inbound logistics consists of raw materials
includes warehouses and multi-ranked suppliers as well as outside processing items.
spread across the globe to competitively exploit Outbound mainly comprises of distribution
high technology, cheap labor and logistics. to various warehouses across the globe.
For the semiconductor industry, lack of
visibility across the supply chain partners, GLOBAL TRENDS IMPACTING THE
operation constraints, inability to respond to SEMICONDUCTOR SC PLANNING
demand fluctuations, smaller product life cycles The underlying purpose of a supply chain is to
and longer lead times perpetuate in supply and make products available to customers in the right
demand mismatches. quantity, at the right time in the best possible

54
condition in the most cost-effective manner. In
classical supply chain term it is called On Time
In Full (OTIF).
OTIF is affected by demand fluctuations,
manufacturing and supplier constraints and
logistics coordination. In addition, a few other
factors like prevailing labor conditions, labor
efficiency, system comfort of users, weather
conditions and so on also have a bearing
on OTIF.
As already described above, wafers
are lead-time intensive products and therefore
sensitivity of the supply chain and availability
of finished goods are largely dependent on
advanced planning. Though the finished good
does not take a long-lead time, the making of
wafers does and the decoupling point of supply
chain is the wafer and not the finished good.
The main performance measure of
supply chain in the semiconductor industry are
dependable demand, resource utilization, process
yield management, throughput time
wafers, inter organizations transportation facilities
and collaboration with contract manufacturing.
So demand planning, production planning
and scheduling all play a very important role in
attaining and retaining business competitiveness.
Seamless integration between contract
manufacturing and Original Equipment
Manufacturer (OEM) is another important aspect
of the semiconductor supply chain.
As evident from the above, the supply
chain in the semiconductor industry is extremely
complex with various factors driving it. Following
are a few key factors that predominantly affect
the supply chain and its planning processes:
1. Business profile: The industry can be
classified as fab or fab-less. A foundry or
fab is a manufacturing facility where
many devices like integrated circuits
and chips are made for customers. Fab-less
companies on the other hand, specialize in
the design and sale of hardware devices
that are implemented on semiconductor
chips. The planning of supply chain,
therefore, is very different for a fab
and a fables company.
While supply chain planning of
a fab is dependent on demand fluctuation
and resource constraints, that of a fab-less
depends on demand fluctuations and
supplier constraints. In fab-less business,
demand arises out of customers in one
part and from self-owned statistical
forecast on the other. The planning
process is also generally divided into
three categories of products:
1. Low Risk-High Volume
products
2. Medium Risk-Medium
of Volume
3. High Risk-Low Volume
products.
The medium risk products are generally
those where there is a collaboration
existing between supplier, manufacturer
and customer. Often, in case of such
fab-less businesses the customers also
select the suppliers, which adds to the
already existing constraints.
2. Type of product: The semiconductor
product portfolio is broad and is
categorized as analog circuits, digital
circuits, computer peripherals,
microcontrollers, network products and
so on. Each category demands a different
supply chain (SC) planning strategy.
The analog and digital circuit
55
 manufacturers are suppliers to the
consumer electronic makers and thus
the planning process must address
consumer induced demand variations.
This calls for a collaborated
forecasting technique and an agile
material ordering system. It also needs
very careful resource constraint
modeling in the system, careful lot sizing,
cyclic and non-cyclic lot ordering, yield
modeling and a flexible scheduling
system. The system used for integration
should be able to handle large volume
of data. The computer peripherals,
network products and microcontrollers
form a different league of business, which
is directly linked to the hardware market
conditions. The collaboration among
supply chain partners is important in any
of these supply chains.
3. Environmental considerations: In recent
times, environmental considerations are
increasingly becoming a major part
of the material supply chain process.
Several key semiconductor processing
materials are currently under scrutiny
by the Environmental Protection Agency
(EPA) and the European Union (EU). This
becomes a major factor in product life
cycle, yield consideration and phasing-
out the old and phasing-in the new
products. Called the ‘Green Supply
Chain’, it plays dominant role in today’s
globalized supply planning.
4. Inter/ Intra company level of system
integration & hardware: With global
networks of suppliers, vendors and inter-
organizational dependencies, the system
on which each individual company runs
56
needs to be good but more important
is the ability and compatibility to integrate
with other systems. A homogenous system
may not be available with various
customers and vendors but system
compatibility is the key.
5. Level of trust and commitment in supply
chain relation: Successful supply chain
performance is based on a high level
of trust and a strong commitment among
supply chain partners. Effective supply
chain planning based on shared
information and trust among partners
is an essential requirement for successful
supply chain management. A lack of
trust among trading partners often results
in close scrutiny of every transaction
thereby increasing the transaction costs,
productivity, efficiency and effectiveness.
Creating value-added activities with
partners becomes almost impossible
and supply chain tools used to improve
efficiency, effectiveness and productivity
—such as vendor managed inventory
(VMI), cross-docking (CD), and
collaborative forecasting, planning and
replenishment (CFPR) — eventually
become ineffective.
SC PLANNING IN A NON-
COLLABORATIVE SUITE SETUP
Almost every major semiconductor business
uses an ERP package for transactional
processing and planning software to plan its
business. The planning process incorporates
various cycles, including a 12-18 months,
long-range strategic planning process, an
8-10 months tactical sales and operations
planning process, and a short-range 3-4 months
execution process. Semiconductor Business
 Business Planning Capacity Planning Execution Planning

Long Range Demand

Forecast Planning for Supply Chain
Capacity. Planning for Real
Time Execution
Planning.

Demand Planning
for Execution

Supply Chain Supply Chain

Forecast + sales Planning for
Orders Business Capacity
Plan Planning.
MES/ ERP
Transactional
Process.

Supplier
Capacity

Figure 1: SC Planning in a Non-Collaborative Suite Set-up Source: Infosys Research

planning can be split into three main parts: out to arrive at an optimum demand, which
• Strategic in turn is to be put to execution (MRP) in order
• Capacity to minimize lost sales with the finite loading
• Execution. capacity of resources.

Strategic business planning process Strategic planning: Long-range forecasts,

is used as an input to both capacity as well
as execution planning. All types
planning are driven by demand. Demand
planning constitutes an integral part of all three
levels of planning. Demand planning, however,
is different for strategic, capacity planning and
execution planning because of constraints and
variations in the outlook of supply that is being
planned. Simulation on demands is carried
which are used in the strategic plans, are not
of so much based on mathematical techniques
as on knowledge and judgment of various
sales/marketing, supply chain and financial
organizations. Strategic planning basically spells
out how much the company can sell over a long
period of time — over the next one to one and
half years — based on market behavior, new
product launches, and competition.

57
 A demand planning tool in this case
is ideally suited to integrate the various inputs
obtained from across the company and integrate
it with the final aggregated demand plan to
be used for supply planning. The long-range
strategic demand plan then lays the foundation
for sales and operation Plan.
Many items and chips have longer lead
times and are required to be procured/ put to
manufacturing well before the start of the actual
finished goods assembly. This is a key element
in defining demands for the strategic plans. The
first step in the strategic business plan which is
a 12-18 month plan looks at long term forecasts
and brings several business groups together in
order to help them plan in harmony to achieve
the company goals. This plan is generally done
at an aggregated product level and is used to
calculate the total supply, which is converted
to monetary terms (based on cost of items and
resources). It calculates projected inventory at the
end of each month for finished goods and semi-
finished goods and is translated into monetary
terms that is used as KPI by management. The
plan also calculates inventory turns, planned
utilization of resources, gross margins, revenue
targets achieved and so on. This plan helps
decide the capacity enhancement required at
different manufacturing setups, to calculate the
percentage of long term demand that can be
fulfilled by using the existing capacity (OEM)
and to determine which capacity element needs
be to expanded within a timeframe so as to meet
the projected demand. These plans help the
management to carry out cost-benefit analysis
for contract manufacturing and to take decisions
for capacity enhancements for wafer fabs and for
final assembly and hiring framework.
Capacity planning: The finalized long-term
strategic plans act as inputs for sales and
58
operations planning. Capacity planning is aimed
at maintaining demand as per the KPIs decided at
the strategic planning level. It also determines the
capacity adjustment or reconfigurations needed
at aggregated product level when more accurate
demand and capacity information becomes
available. The plan helps reduce the lost sales
and meet demands to maximum possible extent.
For very high demand items this is also used for
customer allocations. SOP plans thus give a vivid
picture of capacity variability, demands that can
be met, manufacturing allocations/ capacity,
supplier capacities and customer allocations. All
these go as inputs for demand maintenance in
Execution planning. Many companies use SOP
plan to lot-start wafer and some critical chips
that have long lead times while a few others use
execution plan.
Execution planning: After SOP plans are
finalized, demand is modeled to daily production
capacity and to optimize supply and minimize
lost sales. Execution plan plays the role of an
MRP in the over all manufacturing because it
involves the following:
• Inter manufacturing site transfer of raw
and semi-finished goods
• Planning/ release of actual manufacturing
work orders for execution system
• Planning/ release of requisitions and
purchase orders from suppliers
• Planning machine maintenance and
resource unavailability
• Maintaining safety stock at different
logical stock points in the supply chain.
In effect, it can be termed an MRP plan as it
generates work orders, requisitions and also
inter-department transfers and firming the
generated supplies. Execution planning thus
interfaces with transactional modules such as
costing, order entry, inventory control, bill of
materials (BOM), purchasing, manufacturing
floor control, scheduling and invoicing.
The plan also absorbs the maximum
change in demand due to additions and
cancellations of orders by customers. It is used to
determine the input quantities for each lot at
various semi-finished/ finished goods level.
One important characteristic of
semiconductor planning is that both product
demands and manufacturing capacity are
sources of uncertainty. New manufacturing
processes create high variability in the
yields and consequently uncertainty in the
manufacturing throughput, which in turn leads
to uncertainty in capacity estimation. So the
capacity estimation as done in SOP plan may
not remain the same during execution. Due
to this variability, manufacturing planning
at execution stage also becomes critical to get
the desired benefit of SOP and strategic plans.
A majority of the companies use daily
and weekly plans for execution purpose as these
are also used for finished goods shipment
planning. SOP plans determine the lot to be
expedited to minimize lost sales and execution
plan in effect determines the input quantity as
required to meet the desired objective. Projected
available balance (inventory) from this plan is
used to determine the stock-out situations and
prospective lost sales that are then taken care of
in future SOP plans for lot expedition.
Demand management in this execution
plan is very important as this plan takes forecasts
as the input to firm orders. The adjusted demand
from SOP plan to execution plans is called
as consensus demand. Consensus demand is
highly synthesized demand and is in line with
the strategic and SOP plans.
As clearly illustrated above, planning
cycle in the absence of a collaborative suite is
59
highly complicated and requires very close
coordination among many organizations and
departments to get the best-of-the-breed plans to
gain maximum profitability.
SC PLANNING IN A COLLABORATIVE
SUITE SETUP
The semiconductor industry has one of the
longest and most complex manufacturing
processes in the entire business world, yet
semiconductor customers increase pressure
on manufacturers to decrease production time
while keeping costs low. In turn, manufacturers
press their suppliers to have the correct quantity
of necessary materials at the exact moment they
are needed. Because of technological advances
and changes in demand, semiconductor
manufacturers — and their suppliers — need to
implement multi-organizational planning and
real-time collaboration.
Globally, the trend is for a collaborative
approach to reduce planning cycle time,
improve forecast dependability (as we move
for consensus demand planning), and to get
product committments from suppliers and
contract manufacturing (as we move for consensus
supplier/ contract manufacturing capacity).
Collaborative Planning in semiconductor
involves the following steps:
• Strategic Capacity Planning
• Collaborative Demand Planning
• Collaborative Capacity Planning
• Execution Planning.
Strategic capacity planning: Strategic planning
remains the same with or without collaboration
suite. It is a long-term plan for determining the
company’s objective and business directions.
It mainly plans for the two most important
characteristics of semiconductor industry:
wafer demands and rough cut manufacturing
capacity. It helps companies decide the number
of contractual suppliers needed, possible global
sourcing allocations, utilize bottleneck resources,
monitor start/ completion for each item. This
planning cycle serves as an SOP plan as well as
a strategic plan. It takes into account long term
demand and maps it with existing capacity
and forecast the additional capacity/additional
contract manufacturing needed to meet the
business goals. Once finalized, the plan acts as
an input to the collaborative demand and supply
planning process.
Collaborative demand and supply planning:
Planning starts from demand in a collaborative
planning cycle. The initial guidelines are set
by the strategic plan, which are used by sales/
customer representatives to generate a consensus
forecast along with the manufacturer.
Departments such as sales, marketing
and SCM are involved in this process from the
manufacturer’s side that in turn interact with
sales/ customer representatives to firm up
demand. Mutual trust and open communication
across enterprise holds the key to success at
this stage. Once forecasts are finalized, the
data is incorporated into the planning cycles
for consolidation after which the supply plans
are drawn. The plans are shared with contract
manufacturers/ suppliers.
At the next level of collaboration
planning, suppliers and contract manufacturers
commit on the capable supply scenarios. The
supply-demand match is done at this stage,
mismatches are ironed out. Thus the overall
planning cycle is reduced and is more dependable
as customer representatives together with the
manufacturer draw the consensus forecasts
and suppliers/ manufacturers commit on the
availability of goods.
60
Also, the yield variability is catered
to as manufacturers and suppliers give the
capacity commitment. The process overcomes
traditional planning shortcomings and creates
faster and more accurate plans. Once the
collaborative planning process is completed,
demand and supply are finalized and execution
plan is run. The frequency of data sharing
among representatives, suppliers, contract
manufacturers and OEM decides the quality of
planning.
As semiconductor product life cycles
continue to shrink, managing product transitions
and end-of-life events require collaboration
within and across enterprises. When products
reach full maturity, make or scrap decisions must
be made based on inventory availability, market
demand, available capacity and profit margins,
as well as new product launches. Collaborative
planning in all these cases suits really well for
this industry.
Execution planning: Once collaborative demand
and supply capacities are finalized, demand is
modeled to daily production capacity and to
generate maximum supply and minimize lost
sales. As mentioned earlier, actual execution
plan plays the role of an MRP in the overall
manufacturing. The plan is used to determine
the input quantities for each lot at various semi
finished/ finished goods level considering the
yield of the various operations. This generates
the orders for contract manufacturers and other
suppliers. Thus we see that this planning method
is fast and compact and is more reliable than the
previous traditional planning process using a
non-collaborative set up.
Though collaborative planning has
been in existence for long now, technological
advances and internet communication have now
made it more feasible and effective.
 Strategic
Capacity Planning Customer Forecasts

Long Range
Forecast
Collaborative Demand Planning

Actual Execution Planning MES/ ERP

Transactional Process

Supply Chain
Business Plan
Collaborative Supply Planning

Supplier Capacity

Figure 2: Supply Chain Planning with a Collaborative Source: Infosys Research

Planning Suite

Collaborative planning has the potential
to lower working capital, reduce inventory,
forecast dependable demand, reduce cycle time
and improve responsiveness. Following are a
few advantages of collaborative planning:
Phasing out product: In a collaborative planning
scenario phasing out a product is easier as
all parties involved in the supply chain are
constantly communicated of the changes and
their plans can be adjusted thereof. The forecast
can be reduced or nullified and collaborated with
manufacturing division and suppliers about
the intended future. Without collaboration,
the phasing out will be a tedious and costly
process where the manufacturer must use up
the product till the supplier stops the scheduled
manufacturing in his/her own firm.
smoother. New product forecasts are shared with
suppliers and a future supply plan is charted
based on their feedback.
Tempering risk: With well-defined planning
processes, companies are trying to build a lean
and agile supply chain. While currently available
software do not cater to ‘what-if’ analysis,
a collaborative approach can help through
extensive communication between different
stakeholders in a supply chain network to
temper risks and make necessary amendments
to the plans. The need is to go beyond existing
planning domain and react to a potential
problem ahead of time.
Some common gaps: A few issues that software
of future needs to address are:

New product introduction: With a collaborative 1. Variable yield planning: Yield is a very
planning process new product introductions are important aspect of planning in

61
 semiconductor manufacturing. Issues scenarios, the latter has the ability to manage
arise because of variable yield and actual risk better. Given the highly dynamic nature
yield being different than what is being of the industry, collaborative planning is aptly
planned for future. Balancing maximum suited for giving a new supply chain direction
throughput with proper yield to the industry.
consideration is a big challenge.
2. Wrong routing and supplier selection: REFERENCES
Though not a frequently occurring 1. S Jain, et al., Criticalit of Detailed
problem, sometimes this software selects Modeling in Semiconductor Supply
wrong routing and supplier which Chain Simulation, Proceedings of Winter
means selecting a lower ranked route or Simulation Conference, 1999.
supplier for supply generation. 2. P Dwaraknath, et al., Sales and
3. Optimizing manufacturing capacity: Operations Planning Practices at
The challenge is to suggest correct lot Semiconductor Companies, Survey
sizes to reduce cycle time and operation Report, Joint survey by University of
cost by incorporating demand variability, Texas at Dallas and Motorola Inc., 2002.
supplier response and to also maintain 3. M K Sambara, Collaborations for
high utilization of capacity. Semi- a Hi Tech Contract Manufacturer
conductor manufacturing is both a SME Category, White Paper, Wipro
resource as well as operation cost- Technologies Limited. Also available at
intensive. Hence under utilization is a http://www.wipro.com/webpages/
major concern. At the same time, the insights/contractmanufacturerSME.htm.
industry also battles with fast product 4. S Karabuk, David Wu, Coordinating
obsolescence and volatile demand. Strategic Capacity Planning in the
4. Safety stock calculation and Semiconductor Industry, Technical
maintenance: This is required to cushion Report, Lehigh University, 1999.
demand variability. Two most important 5. I G Kwon and T Suh, Factors Affecting the
factors negating it are obsolescence Level of Trust and Commitment in
and demand constraint and fluctuating Supply Chain Relationship, The Journal
demand. It is difficult ascertain a certain of Supply Chain Management:
quantity as safety stock and tedious to A Global Review of Purchasing and
produce the extra quantity. In reality, Supply, May 2004.
inventory is either over stocked or 6. A Schaller, et.al.Customer Driven
completely starved. Operations Management for Supply
Chain, Technical Paper, Motorola Inc., 2007.
CONCLUSION 7. Achieving Worldclass Fabless Planning,
While numerous factors affect the supply chain Technical Paper, Inc. Also available at
efficiency in the semiconductor industry in both http://www.adexa.com/adexa_library/
non-collaborative and collaborative planning library.asp.

62
 SETLabs Briefings
VOL 5 NO 3
Jul-Sept 2007
Can Transportation Costs be
Stemmed Even Before they Occur?
By Suresh Prahlad Bharadwaj, Sumesh George and Hariharan Noorani
Creating better internal visibility
can help companies proactively reduce
transportation costs
S upply Chain execution in the nineties focused
on reducing inventories. While the fixation
on inventory continues, the macro-economic
situation has since undergone a dramatic change
to being one characterized by high oil prices
and high interest rates. Given the fact that
transportation costs and inventory costs pull in
opposite directions, businesses are struggling to
find the right balance between the two.
Transportation cost, estimated at 3-
4% of sales, is an important aspect of logistics
planning and execution for a retailer [1]. What
was a backburner issue until now has made
a comeback recently, thanks to escalating
fuel costs, driver and capacity shortages. For
example, driver attrition is sometimes as high
as 100% because truck driving as a profession
is mostly viewed as a ‘bad’ job. In the United
States, the hours of service regulations have
fuelled driver shortage further aggravating
transportation costs. Consequently, carriers are
63
in a position to cherry pick shippers they want
to work with. Shippers whose processes and
operations are so inefficient as to increase the
non-productive waiting time for the driver are
usually not candidates for repeat business as far
as the carrier is concerned.
How can retailers react to logistics costs
that tend to erode their already thin margins?
Collaboration with supply chain partners is
something that is much talked about. However,
such external collaboration is often harder to
obtain and even more difficult to sustain. The
often overlooked aspect of collaboration is one
that is internal to the organization. Are there
opportunities that companies are letting pass
simply because they have no internal visibility?
How can internal visibility help make better
ordering decisions while maximizing utilization
of company owned resources. While these
appear to be classical problems dogging supply
chain executives, does the current economic and
 2000 2001 2002 2003 2004
Total Logistics costs as a % of sales 9.44% 9.17% 7.65% 7.52% 8.37%

Component Costs

Transportation costs 37% 47% 43% 35% 39%

Warehousing costs 25% 20% 26% 21% 23%

Ordering / Admin costs 17% 11% 10% 9% 10%

Inventory carrying costs 21% 22% 21% 35% 28%

Figure 1: Logistics Cost Source: Herbert W. Davis Co

business environment make a stronger business visibility coupled with collaboration can
case than ever before? also go a long way in alleviating the problem.

IDENTIFY THE ROOT CAUSE CHANGE ‘INVENTORY IS MY NO.1

In the past, traditional approaches have
focused more on the symptoms than the root
cause(s) of the problem. For example, viewing
transportation as an issue may lead us to
implementing a Transportation Management
System (TMS), one that we expect to act as a
panacea to the creeping transportation costs.
While any decent TMS would streamline your
transportation procurement and generate
optimal loads and routes, it still is constrained
by the inputs provided to it.
The purpose of this paper is to look
beyond the symptoms and at underlying factors
that contribute to the transportation costs.
While well known reasons for a rise in
transportation spend are already highlighted,
the seemingly innocuous reasons often
pertain to organizational behavior. What we
mean by this is that while external factors
such as visibility may be key elements to
running an efficient supply chain, internal
ENEMY’ MINDSET
The late 20th century saw the introduction of
several new concepts such as just-in-time and
quick response which focused on reducing
inventories. This unrelenting focus on reducing
inventories resulted in smaller order quantities
which in turn meant more frequent less-than-
truckload (LTL) shipments. Although this meant
increased transportation spend, the prevailing
low oil prices and low interest rates made it
convenient to ignore the same.
However, rising interest rates and
historically high oil prices have now compounded
the adverse impact of reduced inventories on
transportation costs. Figure 1 depicts the break
up of logistics costs as a percentage of total
sales. It also indicates that transportation costs
which make up nearly 40% of all logistics costs
are steadily increasing. In fact research shows
that, even today, nearly two-thirds of the three
million trucks in the United States are a part of a

64
dedicated fleet [2], a good majority of which are
deployed to carry LTL shipments.
INBOUND TRANSPORTATION
Most retailers consider outbound
distribution as being within their ambit while
inbound transportation as the responsibility of
their suppliers. With inbound transportation
being outsourced to the vendor, it has resulted in
buyers negotiating item rates that are typically
inclusive of transportation costs to
receiving Distribution Centers
Consequently, while the cost of goods sold
represents the landed cost at the retailer
warehouse, it also represents an element of
Retailers must realize that inbound
transportation is as much their responsibility as is
outbound store distribution
trust on the supplier’s ability to negotiate the
best possible rates with carrier companies that
transport the goods for them.
EXTERNAL VISIBILITY
External visibility (typically characterized by
exchange of information through EDI) has
been difficult to achieve given the costs and the
scale of coordination required between trading
partners. Few companies have a full-fledged
implementation of an Advance Ship Notice
(ASN) program covering all vendors and all
items they sell. It seems to be mostly confined to
few large players who have the wherewithal to
invest in EDI and such other infrastructure that
tend to be expensive. Supplier portals, which
can be a relatively inexpensive alternative to EDI,
have seen slower, though encouraging adoption
rates. Consequently, most retailer warehouses
store still grapple with poor external visibility and
therefore work in a reactive mode.
INTERNAL VISIBILITY
External visibility initiatives are inherently
harder to drive and accomplish because they
their have to deal with entities that are alien to an
(DCs). organization with values that are quite different
from that of its own.
However, what is surprising is that in
many instances, departments or business units
within the same organization end up working
in silos for reasons that go beyond the intended
goal of focusing on its profit center.
For example, in many cases it is likely
that multiple buyers’ source from vendors in
the same vicinity. However, for want of internal
visibility, they are unable to see if they could
potentially share a truck and hence the costs.
While the ostensible reason in most cases is that
the ship dates are too far apart or limitations
of the equipment, the end result is the same
– potentially having two vendors who are
located in close proximity of each other deliver
independent LTL’s to the same destination
warehouse. All things being equal, having
65
internal visibility could have possibly allowed
the buyers to consolidate their respective LTL
shipments into one TL shipment at the time of
planning for these PO’s.
Therefore in order for an internal
visibility initiative to be successful, it is imperative
that there be executive level intervention which
is backed by technology capable of breaking such
silos and allowing for seamless communication
between them.
HOW CAN A RETAILER ENABLE
INTERNAL COLLABORATION?
Our approach to transportation costs in a US
retail context has been to first validate our
assumptions via analyst interviews and reports.
Establishing size of the shipment opportunity and
creating internal visibility are key to enabling internal
collaboration at the retailer’s end
Gartner’s January 2006 report confirms the
impact of transportation cost on the overall
logistics cost. It further states that rising
transportation costs threaten to negate supply
chain cost reductions achieved thus far [3]. The
same report also concludes that transportation
budgets worldwide could increase as much as
25% over the next few years. Another AMR
report states that inbound logistics collaboration
technology can enable retailers to reduce freight
budgets by up to 10%[4].
The Grocery Manufacturer’s
Association’s (GMA) 2005 Logistics Survey
had very interesting observations. It showed
66
that while 83 percent of their respondents
cited reduction in transportation costs as their
topmost priority, it also implied that 38 percent
of the transportation costs are due to inbound
transportation [5]. Finally, it is observed that per
mile costs had gone up by over 23 percent in a
two year period.
The first of our two-step approach is to
analyze historical shipments for potential savings
that were foregone for want of consolidation.
This is called the ‘diagnostics’ step. This
step is necessary to understand and quantify
the extent of the problem. The second step is
to analyze future shipments and proactively
identify those that have potential for
consolidation.
STEP 1 – ESTABLISH THE SIZE OF THE
OPPORTUNITY
For this purpose, we begin with processing
historical inbound shipment data. The data
would ideally represent between 6-12 months of
shipment information to reasonably account for
seasonal demand variation.
Using origin zip code and ship date,
inbound shipments are clustered by size,
compatibility and load types, based on spatial
and temporal proximity parameters [Fig. 2].
The spatial-proximity parameter
implies, searching for other inbound shipments
that are potential candidates for consolidation
 Existing Inbound Optimized Inbound

LTL3

LTL2
LTL4 Tour-1
LTL6
LTL5
LTL1
LTL7

LTL8
LTL9

LTL10 Tour-2
LTL11
LTL12

Vendor Warehouse

Figure 2: Inbound Consolidation Source: Infosys Research

within a specified radius of a given
origin-shipping location. The temporal-proximity
parameter described here extends the spatial
search by looking for other inbound shipments
within a specific timeframe, prior to or following
the inbound shipment under consideration.
This further enhances the cluster size, thereby,
increasing consolidation opportunities.
The process would further check for
capacity constraints on the trailer by weight,
volume or pallets and compatibility of product
categories to maximize the creation of full
truck loads (TL). Full truck loads thus built
would then be compared against their
component less-than-truckloads (LTLs) for
freight rates to calculate transportation costs
and hence the savings. Such consolidation
when combined with backhaul opportunities
would further enhance the savings potential.
STEP 2 – CREATE INTERNAL VISIBILITY
AND ENABLE COLLABORATION
Once the potential for savings has been
established, the solution proposes to implement
a similar process for future orders. For those
vendors with a good on-time delivery record,
the process can be performed upstream during
the planning phase. A dashboard-like tool can
provide internal visibility to purchase orders
across categories. Buyers can conduct ‘what
if’ analyses of orders by varying spatial and
temporal parameters on vendor locations

67
 S.No Signal Suitability
1. Appointment Schedules Excellent
2. ASN Very Good
3. PO Good
Table 1: Signal Suitability & Usability
and then adjusting proposed delivery dates.
This process will highlight orders that are
potential candidates for consolidation.
While it is desirable to perform
consolidation using near real-time signals (such
as an appointment made or an ASN), reacting
to such signals can be quite impractical under
certain circumstances. Still, it is important that
due diligence be done. Therefore various signals
such as appointment schedules, ASNs, POs etc.,
should be duly considered in order of suitability.
While analyzing these signals for usability, it is
necessary to first consider the signals closest to
execution (such as appointment schedules) and
then work with ones further upstream (such
as ASN, POs etc.) in the planning process, as
depicted in Table 1.
WHAT IS SO UNIQUE ABOUT THIS
SOLUTION?
A frequently asked question relates to how
the proposed approach differs from what
leading demand planning tools or TMS packages
offer? Most demand planning software take
granular forecast or demand signals from a
single vendor ship-from location and build
truck loads out of it. In fact, using pull-ahead
features, it attempts to top off LTLs into full
68
Usability
Too late to react
Can be too late if generated
after truck leaves origin
Allows for sufficient
time to plan & execute
consolidation
Source: Infosys Research
truck loads by absorbing future demand.
However, those packages may limit such
functionality by item, category or shipping
lane and therefore do not address the divide
amongst buyers from different departments.
TMS packages on the other hand,
take confirmed transport orders as their input
and perform consolidation based on temporal
and spatial search somewhat similar to what is
described in this paper. Spatial search allows
consolidation across multiple stops at the time of
pick up. However, temporal search is restricted
to pick-up windows at vendor locations and
delivery windows at destination DC. Again the
TMS neither has the ability nor the mandate to
proactively make an impact further upstream
in the planning process. By the time the TMS
receives the PO’s, the buyers and vendors
are mostly too far committed to be able to
accommodate changes for the sake of efficiency.
This leaves TMS with the opportunity to
optimize against the constraints of the execution
process alone because it is far too late to impact
the planning process.
Consolidation based on user defined
temporal and spatial parameters, when done
early enough in the PO lifecycle provides the
much required internal visibility to buyers across
departments and help them make an informed
decision regarding collaboration on delivery
dates.
A review of commercial off-the-shelf
applications in the supply chain domain reveals
that none of those applications address the issue
at its roots as proposed here.
HOW DOES DELIVERY DATE CHANGE
IMPACT ON-HAND INVENTORY?
The only argument against making PO delivery
date changes could stem from the likely
adverse impact it has on inventory levels at the
receiving warehouse. However the solution
takes this into account by not allowing the
Holding stock buffers or lowering temporal hold limits
help in absorbing the impact of delivery date changes on
inventory levels
ship date to change on those PO records that
have been identified as ‘must-go’ shipments
by the Demand Planning System. Such impact
can also be minimized by one of the following
factors: allowing for sufficient safety stock
at the receiving DC or by specifying a smaller
temporal parameter i.e., lower days limit
values. In either case, any inventory risk is
expected to be limited to the first cycle of
optimization since future orders will already
be attenuated by the shifted delivery dates and
corresponding inventory levels. The benefit of
transportation savings is also expected to far
outweigh the short-term inventory exposure
associated with this process.
69
CONCLUSION
We have approached the problem based on
the premise that the best chance for reducing
inbound transportation costs lies with tackling
the problem at its root. The fact that the solution
footprint is limited to the retail organization
with little or no expectation from its supply
chain partners makes it a highly implement-able
solution. The solution does not duplicate the
functionality of a TMS; rather it enhances it by
providing a pre-optimized dataset that enables
the TMS to run more efficiently. It is expected to
have minimal investment and a relatively short
implementation cycle with potential savings
realization coming immediately after that. The
fact that the process is minimally invasive requires
little change management since most purchase
orders are typically verified by the buyers prior
to being dispatched to the suppliers. And finally,
the fact that inbound consolidation works even
better when such transportation is controlled
by the retailer causes a transformational shift in
their inbound logistics business, thereby, giving
them a better handle over controlling costs.
REFERENCES
1. Herbert W Davis & Co, Annual Logistics
Survey 2000-2004.
2. H Donald Ratliff, Introduction to
Transportation Planning, The Logistics
 Institute, Georgia Institute of Technology, Research ID Number G001335500.
March 2006. 4. Greg Aimi, Retailer save Money by
3. C Dwight Klappich, Higher Freight Costs Controlling Inbound Logistics, AMR
Increase Need for Transportation Research Alert, October 25, 2005.
Management Solutions, Gartner 5. The GMA 2005 Logistics Survey, IBM
Publication Date: 11 January 2006 Business Consulting Services.

70
 SETLabs Briefings
VOL 5 NO 3
Jul-Sept 2007

Managing Costs by Leveraging

Procurement Information
Intelligently
By Rajib Saha

Augment your enterprise’s spend visibility

and redefine your procurement strategies with the
360° approach to enterprise cost management

C hief procurement officers (CPO) today face

great challenges to comprehend on how
their organizations spend in buying diverse
categories of materials and services. Without a
clear understanding of dollar outflow, it becomes
difficult for them to carve out cost-saving
strategies that can come through improvement
in procurement operations. Enterprise Cost
Management (ECM) can help them understand
their organization spend scenario and develop
strategies to achieve cost saving.
ECM can be defined as the set of
processes, tools and practices to achieve visibility
and control on enterprise spend. ECM provides
insights into areas for savings and exposes
procurement process inefficiencies as well. With
comprehensive view on spend through ECM
organizations can adopt appropriate strategies,
policies and systems to improve procurement
performance.
TWO VIEWS ON PROCUREMENT SPEND
To deliver high-performance in procurement
organizations need: (a) unified view of enterprise
procurement data residing in multiple systems;
(b) multi-dimensional analysis to transform
data into information, and (c) integration of
marketplace, process and technology aspects for
comprehensive insight.
These above-mentioned objectives can
be achieved through a two-pronged approach.
1) an ‘inside-out’ view of transactional spend
data residing across multiple systems through
advanced analytical software applications, and

71
2) an ‘outside-in’ view through category
enrichment, supply market analysis, processes
assessment and technology evaluation.
The ‘inside-out’ view provides visibility
to operational performance whereas the
‘outside-in’ view drives strategies and policies.
Through ‘inside-out’ view, by using spend
analysis applications, one gets to know from
historic data, how an organization has spent for
a given past period. Deficiencies, if any, in the
internal processes can be identified as a result of
spend analysis. These analyses certainly lead to
savings by addressing the identified anomalies.
But they have limitations in unlocking
“Inside-out” and “Outside-in” views of procurement help
enterprises have a 360° understanding of their cost
management landscape
continuous value once the process anomalies
are addressed. Therefore questions arise on –
• How to achieve incremental benefits?
• How to leverage changes happening
within the business context to impact
procurement performance?
The ‘outside-in’ analysis provide insights on
the ‘how’ or ‘what next’ aspects of improvements
by integrating external elements that are
relevant to procurement. This analysis identifies
incremental savings opportunities available
to the enterprise due to changes in external
environment, like supply market, input material
price changes etc.
72
The 360° approach to ECM would be
to follow a gradual approach to understanding
enterprise spend landscape with the ‘inside-out’
view followed by ‘outside-in’ view. The 360°
approach to ECM can help in understanding
end-to-end enterprise spending through the
combination of both side views. In the 360°
approach, outcome from inside data analysis
would be enriched with market, process and
technology analysis to make better decisions.
Before elaborating both views of ECM and
comprehend 360-degree approach, one need to
also understand cost management challenges
that today’s procurement department faces.
COST MANAGEMENT CHALLENGES
Organizational focus on procurement has
radically altered over the last decade.
Procurement has transformed from back office
transaction processing department to a strategic
function. Today, procurement departments play
significant role in organizations in achieving
competitive positioning and price leadership
in the marketplace. Procurement executives
are under significant pressure to contribute to
their organization bottom-line. In industries
like automotive or discrete manufacturing,
where product cost would be a prime factor
for competitive positioning, procurement
function has emerged as strategic differentiator.
To maintain profit margins, the need for high
performance in procurement would be obvious
in such industries.
Procurement executives face a for
number of challenges while spearheading
procurement transformation initiatives. Poor
view of enterprise spend would be a key issue.
This would be due to dispersion of data in
multiple systems across the enterprise. Poor
data quality and inconsistencies add further
complexities. Without proper assessment
decision-making becomes difficult. Traditional
ERP or legacy systems do not provide the kind
of reports that today’s procurement executives
need. Integrating market knowledge to the
internal analysis to enhance sourcing decisions
ECM program enables enterprises
overcome these challenges and draw a roadmap
a holistic performance improvement.
The roadmap would be the outcome of deep
analysis across multiple dimensions. ECM paves
the path for CPOs to meet their cost savings
targets and contribute to their organization
bottom-line.
SPEND ANALYSIS – THE ‘INSIDE OUT’
VIEW
Spend analysis application enables enterprises to
extract, enrich and analyze spend or procurement
data. It allows enterprises understand spend
patterns (by suppliers, categories, business units,

Spend analysis applications bring with them multiple

benefits — primary among them being, getting insights on
savings opportunities and process improvement areas

would be another challenge. There are issues regions) and spend leakages (non-compliance,
with non-standard procurement processes and price variance) happening within existing
technologies among multiple units and regions processes. It provides insights for identification
of an enterprise. of savings opportunities while also indicating at
There are concerns regarding process improvement areas.
introduction of new vendors or switching away Spend analysis application can be
from incumbent vendors, even if that reduces described as a smart data warehouse with
cost. Then there are, other questions like – advanced data cleansing, classification and
• Which technology platforms or point analysis tools. Some of the popular application
solutions are to be given priority? providers in this space are Ariba, Zycus, Ketera,
• What are the market changes in Verticalnet, The Buying Triangle, etc. The process
technology landscape that can be of spend analysis happens in three stages –
leveraged? (1) extract (2) enrich, and (3) analyze.
• Would procurement outsourcing be the Extract – Spend data resides in
right strategy? multiple systems like ERP, payment systems,

73
 Spend data sources across enterprise
Data Source ‘n’
Extract Common
Data
Warehouse
Data Source 2
Data Source 1
Figure 1: Spend Analysis – Extract, Enrich, Analyze for
‘Inside-Out’ View
e-procurement applications, procurement cards
and T&E systems. More often than not, these
systems are built over a period of time and
they are not integrated. The data formats are
different in these source systems. Moreover, an
enterprise can have multiple units or regions
with different type of applications to store
spend data. To get any visibility through such
dispersed data, pulling and storing them in
one single place in a standard form would be
mandatory. Spend analysis application extracts
all instances of spend data across different
systems within the enterprise into its common
data warehouse. Most commercially available
products (COTS) have ready interfaces to
extract data from popular or standard
applications and have tools for easy mapping of
data formats of other systems to their common
warehouse format.
Spend Analysis Application
Supplier Spend Analysis
Reports
Region
Classify
Category
Normalize
Enrich Analyze
Source: Infosys Research
Enrich – Once spend data from all dispersed
systems are loaded into the common warehouse,
cleansing, normalization and classification
would be needed to make the data meaningful.
Such data can be then used for further analysis.
Normally data residing in multiple enterprise
systems are inconsistent. Different systems or
business units may follow different nomenclature
and coding system for the same line item. For
example, ‘Desktop Computer’ may be stored as
‘Computer’ or ‘PC’ or ‘Personal Computer’ in
different systems. Different units or systems may
follow some sort of product classification but
they may not be standard across the enterprise.
From analysis standpoint, unless data is properly
classified and grouped, aggregation would not
be possible. So understanding total spend for a
line item for leveraged sourcing would not be
possible. Neither can price variance be identified
74
for a single line item bought by different units.
Another example of data inconsistency can be
supplier-naming discrepancy. Without a central
master data management process and formal data
governance structure, different business units
within an organization may register the same
supplier in different ways. For example, ‘Hewlett
Packard’ may be stored as ‘HP’, ‘HP Inc’, ‘H.P.’,
‘Hwlt Pckrd’ or ‘Compaq’. It may be possible
that the same supplier ‘Hewlett Packard’ may
have been registered more than once (more than
one supplier code) due to different addresses or
may be registered under different GL codes (as
IT product supplier, as IT service provider, as IT
consultant etc) due to multiple line of products/
services supplies. Unless supplier data have
been normalized, organization cannot estimate
supplier relationship value (total spend) for
better bargains.
Spend analysis application has
capability to add taxonomy information to each
record to arrive at some standard commodity
structure for all products and services. Most
COTS applications have artificial intelligence
based algorithms -- like, Bayesian analysis, K-
nearest neighbor analysis, natural language
parsing and rich knowledge libraries. This
enables granular classification aligned to various
standard commodity taxonomies (like UNSPSC,
SIC, eClass etc) or any custom taxonomy. Spend
analysis applications have rule-based engines
to normalize data inconsistencies like spelling
errors, acronyms, similar descriptions, duplicate
records, etc.
Also, links among suppliers can be
identified to group them, normalize supplier
data and recognize supplier families. Some
niche spend analysis packages have high-
end capabilities (web crawlers) to integrate
information from external sources, like supplier
catalogues and content providing sites with
75
existing suppliers data. Such capabilities allow
enriching supplier data and keeping supplier
related information auto-refreshed.
Analyze – In the previous two stages spend
analysis application extract all enterprise
spend data, integrate them and add necessary
intelligence to them for meaningful analysis. In
the final stage, the analysis capabilities (OLAP
tools) of spend analysis application slice and dice
the aggregated data across multiple dimensions,
viz., category, supplier, region etc. The tools
provide reports with in-depth insights on how
an enterprise have been spending dollars in
different categories with multiple suppliers in
different units or regions. With such granular
visibility, enterprises can perform the next level
of analysis and roadmap planning to save money
directly by negotiating prices with supplier
or indirectly by improving spending related
processes. Simply put, one can get answers to
some key cost management questions like:
• How is the enterprise spending
distribution by dimensions like categories,
suppliers, units, regions etc?
• How is the Pareto analysis by category, by
suppliers? Where to focus in strategic
sourcing efforts?
• What is the aggregated enterprise
spending for any category? What is the
leverage opportunity?
• Who the key suppliers are? How much
is the aggregated spending with these key
suppliers? What is the leverage
opportunity?
• Are there purchase price variations for
any category? What is the savings
opportunity?
• What is the contract compliance
percentage for any category? How much
 spend is through non-preferred suppliers?
• Which units/regions are not following
contracts? Which categories have high
non-compliance?
• Which categories have supplier
consolidation opportunities? Are parts
rationalization opportunities available?
LIMITATIONS OF ‘INSIDE OUT’ VIEW
Reports from spend analysis application
provides visibility to understand what is
happening within the procurement department
of an enterprise. They also bring out intelligence
residing within procurement related transaction
data. Procurement executives can understand
which departments/ buyers have not followed
which corporate contracts. They may find that
some suppliers are supplying same items to
different units (within enterprise) at different
prices.
Through these analyses spend analysis
reports provide great insights for saving money
and improving cost management metrics. But
to take a more holistic approach one needs to
know how organizations can reduce incidences
of spend leakages by changing or improving
processes. It is required to know what
technology up-gradation would be needed in
procurement portfolio to overcome barriers of
manual processing. Such questions would not
be answered by spend analysis application.
Spend analysis application cannot take
into context the effects of changing external
environment. For example, changes in prices
of crude oil would have an impact on spend of
packaging supplies (laminates). The effect on
laminates’ prices would be significant for any
CPG manufacturer. The spend analysis tool
cannot provide such information. From spend
analysis reports, one may discover that within
the enterprise some suppliers are receiving
76
online payment from a few business units while
others are still continuing paper based payment
for those same supplier(s). To correct such
anomalies it would be imperative to know the
process in both scenarios to identify optimum
‘to-be’ process for payment to such vendors.
Such process related views are not available
through spend analysis.
Some of the key procurement areas
that need visibility beyond spend analysis
applications are
• Categories amenable for strategic sourcing
and their risks-benefits scenarios
• Process changes required to improve
performance and implement compliances
• Metrics definition to track process
performance
• Technology up-gradation and outsourcing
opportunities.
If one tries to create the roadmap
from spend analysis reports, it would be based
only on internal information. But that would
limit better business decisions and would not
help realize greater benefits of enterprise cost
management. Strategies made without taking the
external view would not reflect the marketplace
opportunities.
THE ‘OUTSIDE IN’ VIEW DRIVES
‘WHAT NEXT’
To identify incremental opportunities and to
deliver high performance, enrichment of spend
analysis reports would be needed by integrating
other analyses. Without integrating external or
‘outside-in’ view with available spend analysis
reports, ECM programs would not yield great
returns. Figure 2 depicts on how after such
integration, the ‘outside-in’ view drives ‘what
next’ in terms of spelling out spend management
 Supply Market Intelligence
Spend Analysis
Reports
Spend Management Strategy & Roadmap
Figure 2: ‘Outside-In’ View Drives ‘What Next’
roadmap. Enrichment through ‘outside- in’
analysis happens in the following dimensions –
• Supply Market & Category
• Process & Benchmarking
• Technology & Outsourcing.
Supply market & category – Spend analysis
provides granular visibility on dollar outflows
made in different categories in the past period
for the enterprise. What would be important
from an implementation plan perspective is the
77
Category Information
Enrichment
Process Analysis
Metrics Benchmarking
Technology and
Outsourcing Assessment
Source: Infosys Research
identification and prioritization of categories for
strategic sourcing. The spend analysis reports
would not provide clear direction on this. Two
factors drive the decisions for strategic sourcing:
a) estimated savings opportunities, and b)
category risk
‘Estimated savings opportunities’ here refers
to the savings identified after enrichment of
spend analysis reports based on market facts.
That would not be just the opportunity value
(dollar value) as an outcome of spend analysis
application but also properly validated for
possibilities in actual business context. Or
the opportunity may be identified only after
integrating market information. For example,
for certain categories, the input prices may have
changed over the past few months and those
may affect costs of those categories. Technology
improvements could result in changes in prices
of certain materials like IT hardware. Such
market information needs to be factored in while
evaluating savings. Or initiating low-cost country
sourcing for some categories may emerge as a
lucrative option. China or India has emerged
as preferred sourcing destinations for many
categories like electrical accessories, electronics
parts and small-size automobile components. So
what can be the impact of sourcing from China
or India? Reports from spend analysis enriched
through external information on supplier markets
and/ or category related information can help
determine ‘estimated savings opportunities’ for
different categories.
Finding ‘estimated savings
opportunities’ would not be sufficient for decision
on strategic sourcing. The risk associated with
categories needs to be examined and estimated.
To assess risk of supplier change or new supplier
introduction through strategic sourcing,
understanding of supplier market dynamics
would be crucial. The spend analysis application
may show, through supplier ‘grouping’ and
‘parent-child’ linking, that significant amount
of spend would be possible through establishing
one supplier across the enterprise for some
categories (single vendor scenario). That
would depict as opportunities for volume
leverage and savings. However, while assessing
category risk one may discover that those items
are proprietary in nature and the leverage would
not be feasible for cost-savings. While category
risk assessment is done, one should assess
areas like:
78
• Supply disruption possibilities
• Vendor switching costs
• Business impact on failure
• Technical complexities
• Internal organizational resistance and
change management issues
• Commitment with incumbent suppliers
• Supplier industry competitiveness.
From rapid execution standpoint, it
would be prudent to take template-based
approach for assessing ‘estimated savings
opportunities and ‘category risk’. Otherwise
it would be difficult to capture the right
information in a structured format. Template
driven process programs can make the process
of enrichment and subsequent assessment
standard and repetitive. Once estimation of
both parameters is complete, one can create a
‘category matrix’ based on ‘estimated savings
opportunity’ on one axis and ‘category risk’
on the other. The matrix helps prioritize
categories and phase-out strategic sourcing
implementation plan.
Process & benchmarking – Processes are core
to metrics improvement within any business
function and procurement is no exception.
Process views are essential to relate findings
of spend analysis to process deficiencies. For
example, maverick spending or non-compliance
to contracts is the result of non-existence of
processes to integrate contract prices to e-
procurement systems. To improve on metrics or
to implement compliance requirements, process
analysis exposes gaps to plan for process re-
engineering. For a comprehensive process view,
it would be essential for process owners to get
the ‘as-is’ process maps across business units
for different category groups (direct, indirect,
MRO materials and services procurement) and
functionalities (requisition, procure-to-pay,
contract management, supplier registration,
supplier collaboration and supplier performance
management). Such process analysis allows
identifying best practices within the enterprise
and plan for enterprise process harmonization.
However, harmonization would be easier
said than done. There are multiple challenges
ranging from system variations to internal
people issues.
The other side of process view is
defining and tracking of key metrics that can be
rolled up to measure procurement performance.
Unless the existing performance can be
measured and benchmarking metrics can be
set, planning would never be objective. As a
result of improper planning, implementation
would falter. Metrics for benchmarking need to
be defined at three different levels that address
metrics view requirements of different users
within the organization -
1) Executive - For example, total spend as
% of sales, no. of suppliers per billion
dollar of spend, procurement operations
cost per billion dollar spend, no. of FTE
per billion dollar spend, customer
satisfaction rating, etc.
2) Operational – For example, cost reduction
as a percentage of spend, % of suppliers
that consume 80% of spend, % of spend
under procurement, cost per purchase
order, % spend through strategic
sourcing, % spend through
eProcurement system, etc.
3) Process – For example, requisition to
PO cycle time, PO to settle cycle time,
contract compliance %, % of contracts
negotiated centrally, % late payments
to suppliers, no. of POs processed
per FTE, etc.
79
While defining procurement metrics, linking
them across levels needs to be kept in mind. That
enables procurement performance measurement
by rolling up as well performance improvements
through cumulative impacts.
Technology & outsourcing – enterprise
procurement technology landscape is
fragmented. On the one side are traditional ERP
or legacy systems to maintain transaction and
master data and on the other, there are niche
applications for specific functionalities (like
contract management, order collaboration, e-
procurement, catalog buying etc.) or specific
category management (like travel & expenses,
print, contract labor etc). Then there are a host
of standards for supplier connectivity,
collaboration and architecture viz., EDI,
Rossettanet, XML, SOA.
Outsourcing is rapidly growing in
the procurement space. For industries, like
automotive or industrial manufacturing,
procurement can be a core competency area
and strategic to operations. For many others,
like banking or financial services, procurement
would be non-core and a support function. So it
may be feasible to transition entire procurement
function to outsourcing service provider(s) in
certain industries while it would be limited
to specific processes or categories in other
verticals.
Technology and outsourcing assessment
provide insights for technology implementation
roadmap. To start with, one needs to understand
the existing application portfolio, IT architecture
and standards used by the enterprise. The ‘as-
is’ system landscape blueprint helps identify
bottlenecks that need technology intervention.
For technology requirement analysis,
a step-by-step approach can be followed.
First, requirement of solutions that meet the
organization’s priorities can be ascertained
and application providers’ market can be
screened to shortlist vendors. Then high-
level selection criteria template needs to be
defined based on functional and technological
fitment requirements. The packages that match
the criteria can then be evaluated for cost
and proof-of-concept. Thus available
technology options can be evaluated and
baseline plan for technology implementation
can be drafted.
For procurement outsourcing
assessment, it would be better to understand
overall organization plan on outsourcing. Then
it would be prudent to work on creating business
cases for procurement outsourcing. To do so, one
needs to know
• Which processes or categories would be
feasible to outsource? What would be
the risk?
• How would transition take place? What
time frame would be required for steady
state?
• What would be the new process and
organization scenario?
• What would be the cost arbitrage savings,
long-term benefits and ROI?
Answering these questions needs
granular analysis of the ‘as-is’ process map
created earlier, plus a scrutiny of organizational
changes and governance structure. This can be
achieved through a two-pronged approach –
1) Assessment of processes for their fitment
to off shoring/ outsourcing and associated
benefits thereof, and
2) Analysis of organizational and governance
structural changes through frameworks
like RACI.
80
The objective of such assessment would
be to identify outsourcing feasibilities and
development of appropriate business case.
NEED FOR TEMPLATES
Analyses across these multiple dimensions
external to transaction data enable procurement
teams to create a transformation roadmap. They
would be able to make informed decisions on
the initiatives that are to be pursued and hence
plans can be prioritized. The shift from ‘what
happened’ to ‘what next’ and ‘how’ would result
in tangible benefits of ECM. Such shift would
be possible only after enriching spend analysis
reports through these ‘outside-in’ views.
However, driving ECM for long-
term benefits would require template driven
program management. But creating enterprise
wide templates and assessments of widely
diverse categories, processes, technology and
outsourcing require substantial expertise.
The crucial aspect is that these methodologies
and processes must be repeatable. Without
template based data collation and evaluation
process, repeatability cannot be achieved. That
would make the program a one-time affair
and incremental improvements would not be
possible. Organizations would need their
resident subject matter experts to chip-in to create
templates and drive these complex evaluation
processes. In many areas, organizations may need
to engage external procurement consultants.
This would be especially true
especially for MRO (maintenance, repair and
operations items) and indirect procurement
categories (categories that are not used directly
into end product) where organizations may not
have adequate skill set, market knowledge or
resources. In industries like banking or financial
services, such expertise may not reside at all
within the organizations.
BENEFITS OF 360° APPROACH
The 360° approach integrates enterprise
spending insights with other relevant analysis
and external information. By leveraging internal
procurement data with external information
organizations realize two-pronged benefits–
1. Identification of saving opportunities
and strategies to achieve those savings
like – increased strategic sourcing,
aggregation of purchases, supplier
consolidation, price rationalization, part
standardization etc.
2. Identification of areas for improving
process efficiencies by process changes,
technology adoption and outsourcing.
These impact cost metrics and adherence
to contracts and regulatory requirements.
SELF-SERVICE OR MANAGED SERVICE
ECM provides organizations significant gains
that are not easy to ignore. With more and
more companies adopting such programs,
ECM would be a key area for any organization.
However, the important question would be
whether ECM programs can be carried out in-
house or to through managed service route. The
answer lies within the broader organizational
approach to information systems management
and outsourcing. If answers to a majority of
the below-mentioned questions are not in the
affirmative, it would be prudent to take managed
service route or a hybrid approach -
• Is procurement core to enterprise
operation and strategic to competitive
positioning?
• Are in-house experts available to track
and integrate market information to the
majority of categories?
• Can resources be deployed to map
enterprise wide ‘as-is’ process and systems?
81
• Can evaluation of technology solutions
be carried out in-house?
• Are investments in a spend analysis tool
an organizational priority? Would the
ROI in spend analysis investment be
significant and make the investment
attractive?
• Are there resources available to manage
such applications and maintain it?
CONCLUSION
In today’s competitive market, procurement
departments of organizations are under pressure
to deliver high performance, especially, in
cost management metrics. To overcome cost
management challenges it would be imperative
to get comprehensive knowledge about
enterprise dollar outflows and the available
opportunities (internal or external driven) to
control it. The 360-degree approach to cost
management provides complete view of ‘as-is’
procurement state as well as identifies available
opportunities to reduce cost. With combination
of ‘in-side out’ and ‘out-side in’ analysis through
360° approach, organizations can get insights into
their transactional data as well as understand the
effects of external dynamics. That allows them
to adopt appropriate strategies, IT systems and
processes that meet the demands of their cost
management objectives. It would be prudent to
drive ECM through template-driven program
processes so that such programs can be initiated
at regular intervals. Equipped with ‘glass pipe’
view through ECM, enterprises can thus achieve
procurement transformation.
REFERENCES
1. Mickey North Rizza and Lora Cecere,
Successful Supply Analytics: Five Areas
To Know, AMR Research , November 16,
2005.
2. Rip Greenfield, Spend Analysis: The First www.ketera.com/newsletter/best_
Step in Strategic Sourcing, 90th Annual practices_012207_SA.html.
International Conference Proceedings, 4. White Papers: A Practical Take on
Institute for Supply Management, 2005. Spend Analysis , BIQ at http://www.
3. Ketera.com, Spend Analysis – you can’t busiq. com/WhitePapers.html.
live without it, Best Practices in 5. Sunil Chopra and Peter Meindl, Supply
Corporate Procurement, Vol. 1, Issue 1, Chain Management: Strategy, Planning
Jan 2007. Available on http:// and Operation, Prentice Hall, US, 2000.

82
 SETLabs Briefings
VOL 5 NO 3
Jul-Sept 2007

Aligning Treasury Services

Offerings to the Corporate
Financial Supply Chain
By Sabitha Vuppala and Sujata Banerjee

Innovation is the key to surge ahead

in an increasingly commoditized treasury
service business

B anks that want to stay at the top in the FINANCIAL SUPPLY CHAIN
Treasury Management Services business The purchase-to-pay and order-to-cash cycles
need to innovate now more than ever. With constitute views of the trade cycle from the
stagnating revenues and services getting perspective of the buyer and the seller
increasingly commoditized, banks are respectively [Fig. 1]. A majority of companies
undertaking initiatives to provide value added deal with suppliers as well as their customers
services by aligning themselves more closely and therefore need to manage both these
with the corporate client’s financial supply chain. cycles. Accordingly financial supply chain
We examine some of the drivers and trends of management involves looking at the
this industry, and based on this, draw out an end-to-end financial flows across these
agenda of initiatives for banks in this space. processes.
Purchase-to-pay cycle

Select Negoitate Order Receive Process Reconcile

supplier credit terms goods goods and payment payment
invoice

Order-to-cash cycle
BUYER SELLER
Receive Negotiate
Reconcile Collect Send goods Prepare
purchase credit terms
invoice Payment and invoice quote
order

Figure 1: Traditional Financial Supply Chain Processes Source: Adapted from Introduction to Financial Supply Chain
Management, Treasury Today, http://www.treasurytoday.com/

83

Posting Credit
receivables Management
Collections
Cash
Management
Customer
Risk
Service
Management
Accounts
Payroll Payable
Figure 2: Financial Chain Activities
Source: Infosys Research
DRIVERS FOR CHANGE: ENTERPRISES
Significant inefficiencies in this
supply chain offers enterprises the opportunity
to streamline business processes and gain
significant cost improvements. There is a need
for client and trading partner integration and
to move from paper to an electronic form at
all stages of the cycle. There has already been
considerable investments in ERP systems that
help achieve operating efficiencies. But while
the physical supply chain automation is well
underway, there is still plenty of potential for
automating the financial supply chain process.
Enterprises are looking at means to
improve collection cycles, benefit from discounts
through faster payments, automate the manual
processes involved in 3-way reconciliations of
payments, invoices and purchase orders. And of
course, better fraud protection continues to be an
important driver.
There is always an increasing demand
for faster and more accurate information -
be it industrial data or financial data. This is
needed to take the guess work out of managing
working capital, and making it more of a science.
Enterprises can do with all the help they can get
to facilitate liquidity management. The corporate
treasury function within a large and complex
business needs information at its finger tips in
order to act like an internal banker.
All these forces come together to create
the need for enterprises to incorporate payables,
receivables and treasury management internally
within the ERP systems, and with their banks, in
order to integrate functionality. Figure 2 depicts
a state of close integration and synergy among
all activities across the financial value chain.
DRIVERS FOR CHANGE: BANKS
The focus of enterprises to better integrate and
automate the financial supply chain provides
an opportunity for banks. Revenue growth
financial from traditional treasury services has been
stagnating. Increased competitive pressures
continue to strain deposits and fee growth and
squeeze bank’s margins. More companies are
operating globally than ever before, which
is resulting in increasingly complex treasury
processes for a wider range of firms. On one
hand, there is growing disintermediation in
payments as far as banks are concerned while on
the other, there are non-traditional competitors
as well as marketplaces and other intermediaries
coming into the picture with newer innovations.
Also with the consolidation among banks still
continuing, the strong players are getting more
stronger and competition is becoming tougher
for everyone left in the fray.
Industry regulations also continue to
keep banks on their toes – even though they have
always been around, the reputation risk that
non-compliance now poses is a big motivator
for banks to want to keep a step ahead of the
84
 Changing
Industry Dynamics

TREASURY MANAGEMENT SERVICES

– BANK OBJECTIVES
• Meeting long-term strategic objectives
Emerging • Streamline processes Increasing
Technologies • Lower costs Customer Demands
• Reduce risk / fraud losses
• Improve efficiency
• Increase working capital efficiency reduce float)

Heightened
focus on Risk and
Compliance

Figure 3: Factors Impacting Banks’ Objectives Source: Infosys Research

regulators. Thus there has to be a heightened enhanced functionalities in treasury services.

focus on risk and compliance.
Enterprises are increasingly demanding
more cost effective and integrated cash
management “solutions” from their banks, as
opposed to the plain vanilla products in silos.
While enterprises have a desire to consolidate
their banking relationships and regionalize/
centralize their treasury operations, credit
relationships are not a deterrent for them for
looking out to other “better” options.
New and emerging technologies
continue to be an important force for banks.
Migration of checks to electronic payments in the
B2B market is a recognized trend with initiatives
like Check 21 and solutions like Remote Capture
being key facilitators for financial supply chain
integration [3]. There are several mature vendor
solutions available in the market offering
Figure 3 depicts banks’ business objectives and
the forces driving banks towards changing these
objectives.
EMERGING TRENDS AND AGENDA
FOR BANKS
We see several business line trends that have
emerged and are continuing to develop as a
result of the afore-mentioned forces and drivers.
Payments today are a commodity – intense
competition is constantly forcing banks to reduce
prices around payment processing service. There
is however a huge scope for banks to increase
efficiencies, provide greater visibility into the
entire payments process and better integrate
the payments process into the procurement/
supply chain. Banks can, with such value added
services, charge premium and provide a boost to

85
 Sales & Account Payments Receipts Liquidity & Account
Opening Management Management Investments Servicing

• Cross sell • Mobile • Corporate Ac • Multi-bank • Multi-channel

payments counts Global Alerts
• Least Cost Pay Receivable Enterprise- (Txn & Acc
Product ment Routing Management wide Reporting Information)
Innovation • EIPP and Analysis
• Pooling and
• Flexible Product Line Architec ture Netting
• Cash
• Automation
Forecasting
of Account
opening and • Global billing/ Relationship pricing
servicing • Straight • BPO – Outward
–through and Outstation
• Cross Executive processing Clearing
Operational Reporting
Agility Integrated Risk • BPO – Message
& Compliance; Repair, Paper
Predictive Fraud Instruction T1
Mgmt.
• Process Tools – Imaging/Workflow etc

• Intelligent Exception Management

• Shared Services Platform
• Integrated Risk & Compliance: Predictive Fraud Mgmt
Distribution • Customized • SWIFTNET for Corporates
Interfaces
• Integration with Corporate Treasury Systems • Desktop Access
• Integration with Corporate ERP systems Mobile Banking
• Client touch
• Virtual point
Purchasing Integration
cards and
Biometric Tech

• Common clearing / Settlement

Infrastructure
Optimization • Enterprise transaction warehouse
• Enterprise transaction warehouse
• Modular Enter prise Application Platform

Figure 4: Banks Agenda and Specific Initiative Examples Source: Infosys Research
to Support the Financial Supply Chain of an Enterprise

the stagnating revenues and declining margins relationships with employees and vendors. This
in payments. On the payables front, the trend includes technologies such as check truncation
to convert from paper to electronic is ongoing. and imaging. At the same time, the use of
It is more cost effective, safer and improves Automated Clearing House (ACH), wires and

86
purchasing cards is increasing. Banks need to
offer a best-of-breed package of options to suit
different clients.
Another ongoing and rising trend is the
outsourcing of payments. Third-party processors
are increasingly used, particularly for checks, and
ERP. Treasury Management solution vendors
are thus capturing market share as clients look
to increase efficiencies.
On the receivables front too, the
movement from paper to electronic continues
and various initiatives that support reduction
of float are gaining success viz., Accounts
To integrate supply chain management across the
treasury services, banks have to focus on: product
innovation, operational agility, distribution and
infrastructure optimization
Receivable Check (ARC) Conversion, Back Office
Conversion (BOC), image lockbox and so on.
The other imperative is to expedite
information and have remittance data linked
to the notification of incoming collections.
An important step in truly leveraging B2B
electronic commerce is the use of electronic
invoice presentment and payment (EIPP). The
driver for this is to improve speed of invoice
delivery to customers, reduce costs associated
with paper invoicing and reduce costs and
errors associated with processing paper
payments, especially partial payments, disputed
invoices etc.
Outsourcing of receipts management
while clearly an emerging trend with benefit
potential similar to payments outsourcing, is an
87
area still being gauged cautiously by enterprises
given the criticality of managing inflows.
As far as information reporting and
liquidity management initiatives go, the driver
clearly is to improve cash flow forecasting and
make better investment decisions.
Based on the industry drivers and
trends discussed above, we have categorized
the agenda for banks to address these along four
dimensions: Product Innovation, Operational
Agility, Distribution and Infrastructure
Optimization. Figure 4 depicts some important
initiatives that industry leaders are undertaking
across these dimensions, in order to assist
their corporate clients in improving and
integrating financial supply chain management
across the Treasury Services process value
chain.The subsequent sections of the paper
delve into a description of some of the specific
initiatives that align with these agenda items
for banks.
INFRASTRUCTURE OPTIMIZATION:
ENTERPRISE PAYMENT SOLUTIONS
Banks are offering accurate and timely payments
to their customers, with a choice of payment
methods – ACH, wires, checks, cards. While
the individual payment processors are mature,
stable and function-rich, the underlying systems
have evolved as product silos.
 TECHNOLOGY INITIATIVES
• Common Business Services: Centralize
common functionality through reusable
services related to Compliance, Screening,
Billing & Pricing, Risk Mgmt, FX, etc.
• Messaging Hub: Phase out diverse, direct
communication between internal applica
tions and external networks. Use a common
messaging format (XML based).
• Transaction Warehouse: Integrate data
fragmented across multiple databases. Use
common data model (ISO 20022).
• Business Continuity Planning
Figure 5: Technology Initiatives and Business Benefits of
Enterprise Payment Solutions
The resulting infrastructure is a maze
of independent payment systems, databases and
direct, point-to-point connections between front-
end channels and payment networks.
All these independent payment systems
have common functionality, data elements and
use similar software services (risk and fraud
management, AML, currency conversion, pricing
and billing etc.) However, this functionality
currently is duplicated across the different
payment systems. There is a huge opportunity for
banks to consolidate the common functionality
onto a single shared platform, and offer best-of-
breed payments solutions at an enterprise level.
Payment transaction information is
also spread across different databases. Without
a common transaction warehouse, banks cannot
get an enterprise-wide view of payments or
apply analytics on all the transaction data lying
in diverse databases.
Figure 5 summarizes the various
business benefits of Enterprise Payments
solutions as a result of technology initiatives
undertaken.
88
BUSINESS BENEFITS
• Increase operational efficiencies and
agility: Reduced transaction and technol
ogy maintenance costs; increased speed to
market.
• Enhanced reporting and analytics:
Enterprise-wide view of payments
Better tracking of payment transactions
• Offer best-of-breed payment solutions:
Intelligent Routing, Alerting, Rich
remittance information for straight-through
processing.
• Operate round the clock
• Improve Compliance and regulatory
oversight
Source: Infosys Research
Some prerequisites for straight through
processing include:
• Remittance information as part of the
payment transaction in standard formats
supported by banks, enterprises and
payment networks
• Connectivity between bank’s cash
management systems and corporate ERP
systems to allow automated transmission.
Currently there is a lack of system
integration between bank’s corporate cash
management systems and corporate Accounts
Payable and Receivables systems. Companies
must hence execute multi-step manual processes
while both sending and receiving wire transfer
payments. Frequently, wire transfer payments
are sent separately from the remittance
information, creating additional operational
complexity. The receiving party must reconcile
the remittance information to the payment and
post the payment manually to the appropriate
accounting system.
 Banks are responding by collaborating EDI. This is gradually being incorporated as a
with payments network of enterprises in standard in bank/third party offerings [5].
defining and promoting the use of messaging and Wire transfers offer the advantage of
remittance information standards. same day remittance and more secure payments.
The focus of internal technology However, without standard remittance
initiatives is to build an enterprise payment standards, wire transfers lead to inefficient and
solution platform that will feature: manual reconciliation. Currently the Clearing
House and Federal Reserve are finalizing the
• Emerging payment transaction data and choice of standard for Wire Transfers – between
messaging standards (XML based EPN EPN STP 820 and ISO 200022 [5].
STP 820and ISO 200022) that enable While EDI today supports sending and
capture and transmission of rich receiving payments and remittance information
remittance information electronically, it is expensive, not universally
• Sophisticated data transformation and accessible, and very platform specific. Large
connectivity tools and that allow banks corporations are increasingly demanding XML
to receive and deliver payment based standards. ISO 20022 promises to be the

Operational agility can be bolstered

with the convergence of multiple payment standards
in the long run

information as required by enterprises universal XML based payments standard. The

(XCEL, pdf, proprietary formats, XML
etc) through various channels (web,
email, file etc).
OPERATIONAL AGILITY
Emerging Remittance Information Standards
ACH transactions carry relatively
remittance information than wires and enable
faster and more automated reconciliation of
incoming payments with invoices. The EPN
820, an EDI lite remittance standard supported
by NACHA, allows companies that are not EDI
enabled to more easily and inexpensively use
goal of ISO 20022 is to enable a convergence of
multiple payments standards into one in the
long run [5]. In the short run, it facilitates the
coexistence of these multiple standards.
PRODUCT INNOVATION
richer Enhanced Reporting and Analytics
A centralized payments transaction warehouse
will allow banks to get an enterprise-wide
view of their payments business. The
advantages to banks are two-fold: one is
the business intelligence around volumes,
costs, profitability, growth trends across all

89
payment channels and customer segments.
This business intelligence will enable a bank
to take better decisions around customer
offerings, pricing etc.
Of more direct impact to enterprises,
is when banks use the intelligence to improve
efficiencies in the payments process. For
example, banks can identify if their customers are
making payments to other customers of the
same bank, but through competing banks. Bank
users, in such a not-so-uncommon situation,
would be alerted and could contact their customers
and beneficiaries. The beneficiaries then can
instruct the corporate to make payments to their
Traditional treasury products no longer bring the moolah.
Accounts payable and receivables offshoring services are
fast emerging as revenue rakers for banks
accounts with the same bank – reducing the time
and cost of the transaction as well as enhancing
liquidity. Enterprises also get the benefit of
greater visibility into the payments transaction
data - with more online and detailed reporting
around payments transaction volumes,
value, fees, cash positions that can be sliced
and diced by region, subsidiary, payment
methods etc. Availability of more accurate and
online information allows the banks to offer
better alerting services to their customers.
Accounts Payable/ Receivable Offshoring
With stagnating revenues from traditional
treasury products, banks are exploring
opportunities in Accounts Payable and
90
Receivables offshoring services. Banks are not
traditionally outsourcing experts, nor do they
have domain expertise in the corporate supply
chain management area. Hence banks will be
offering services that are extensions of their
current Accounts Payable and Receivables
offerings e.g., wholesale lockbox. Examples of
two such value-added services are described
below.
• Inverse receipt and storage: Majority
of the invoices issued today are on
paper. Banks can extend their wholesale
lockbox infrastructure and mailing
logistics success to receiving incoming
invoices at lockbox locations, scanning
the invoices, extracting data (manual
data entry or OCR) and archiving and
imaging the invoices. Payables files are
created from the invoice data and
transmitted to the corporate for further
decisioning on payments.
• Accounts receivables matching: A
few banks have started to offer enterprises
the service of matching incoming
payments against receivables data.
Enterprises share data on receivables as
well as payments received from other
banks to the bank providing this service.
The bank uses intelligent matching rules
(based on invoice amount, date, LIFO,
 FIFO etc.) to apply payments to
receivables. Any mismatches are then
sent back to the enterprises for
manual investigation and decisioning.
With a root cause analysis applied on
recurring exceptions, future exceptions
can be reduced and the process improved.
LIQUIDITY MANAGEMENT
Cash Forecasting And Global Multi-bank
Information Reporting
Cash forecasting is of strategic importance
in managing liquidity, but is still largely
spreadsheet based. The challenge is to obtain
timely and accurate information to feed the
forecast. This is further complicated when
enterprises have global presence and multi-
currency transactions. A cash forecasting solution
therefore must automate receipt of information
from multiple sources while assuring data
integrity. It should have strong analytical
capabilities while being easily customizable to
an individual enterprise’s needs and objectives.
Clearly, near real time multi-bank
reporting is a related need, to provide companies
with updated bank balances across currencies
and countries with different banks. Multibanking
solutions allow companies to work with their
lead cash management bank and integrate
with other banks. Information is aggregated
using SWIFT messages to facilitate account
reconciliation, concentration and liquidity/
investment decisions. It will also allow payment
initiation from third party accounts via electronic
or web-based platforms.
CONCLUSION
There is tremendous scope and value in
streamlining and improving the financial supply
chain processes. Enterprises of different sizes
in different industries must assess their specific
91
objectives and priorities from the general drivers
and trends described.
Similarly banks cannot offer everything
to everyone, but they must innovate to remain
competitive in the increasingly commoditized
treasury services business. The frameworks
developed in this paper could serve as a starting
point to take a holistic view of the business, the
imperatives and decisions on what the banks’
agenda should be to address these imperatives.
REFERENCES
1. Introduction to Financial Supply Chain
Management, Treasurytoday website
at http://www.treasurytoday.com/
publications/articles/2007-01/
2007-01,%20January,CashManageme
nt.pdf.
2. Alenka Grealish, Scaling the e-financial
Supply Chain Mountain, Celent,
May 2004.
3. Business to business wire transfer
Payments: Customer Preferences and
Opportunities for Financial Institutions,
The Clearing House Payments Company
website at http://www.theclearinghouse.
org/docs/002580.pdf.
4. US Enterprises Call for Streamlined
Wire Transfers, gtnews website at http://
www.gtnews.com/article/6554.cfm.
5. Alenka Grealish, Standards: The Pony
Express to E-Payments Part II – Pioneers in
B2B Standards Development and
Dissemination, Celent, December 2006.
6. Bob Meara, AP and AR Outsourcing: Not
for the Faint of Heart, Celent, March 2007.
7. Nancy Atkinson and Eva Weber,Top
Trends in U.S. Wholesale Payments:
Much More Than Processing Efficiency,
Aite Group, November 2006.
8. Douglas McKibben et. al., Research
 Preview: Pressure on Payment Standards, Kun, Payment Hubs Are the First Step in
Gartner, April 2006. the Evolution of the Banking Payment
9. Christophe Uzureau and Maria Luisa Business, Gartner, March 2006.

92
 SETLabs Briefings
VOL 5 NO 3
Jul-Sept 2007
Management of Supplier Risks
in Global Supply Chains
By V. S. Srividhya and Raj Jayaraman
Improve predictability and business
continuity in global supply chains using
the SRM framework
T he US manufacturing sector contributes
over 12% of total US GDP and two-thirds of
total US exports [1]. In recent times, companies
in this sector are faced with significant
operational pressures. With hourly labor costs
in China estimated to be 3-4% that of the US
[2], US manufacturers are finding themselves
getting priced out of the market by lower cost
manufacturers. In addition to the pricing pressure
from overseas, the industry has witnessed an
inflationary trend in raw material and energy
prices. In the last 5 years, for instance, the base
steel prices have almost doubled [3]. Energy
prices, which cause a direct increase to overhead
costs, have also increased.
With all the core inputs to
manufacturing (labor, material, overhead) seeing
tremendous price hikes, manufacturers have
expectedly responded with a relentless focus on
operational cost reduction. The trend towards
low cost country sourcing has been rapidly
growing. A recent study by Boston Consulting
93
Group mentions that LCC imports are growing
at over 20% in certain categories [4]. Spend
aggregation, leveraged sourcing and modular
sourcing - where vendors source not just parts
but entire modules including accompanying
services (design, inventory management, after-
sales service), is also a growing trend.
While these strategies have yielded
significant operational benefits, recent events
have proved that there are also inherent risks
that could significantly impact benefits. As
manufacturers source globally, they increase
their lead-time and vulnerability for supply
chain failures.
Risks impact total cost of ownership
through their effect on quality, delivery and
end customer satisfaction. There is a need to
effectively assess these risks and develop a
comprehensive mitigation approach to ensure
supply chain continuity.
This paper is focused on providing
a framework to: (i) assess the negative impact
of risks in global supply, (ii) applying a supplied radio frequency chips (RFCs)
comprehensive framework to identify and to Ericsson. The company was not
mitigate the supply risks, and (iii) highlighting aware of the supply problems for weeks,
critical success factors and benefits of the by which time its ability to meet
framework. customer demand had been seriously
compromised. And because Ericsson
THE RATIONALE FOR SUPPLY RISK relied exclusively on the Albuquerque
MANAGEMENT plant for the RFCs, they had
In the context of global supply, there are a nowhere else to turn for these
variety of factors that can contribute to supply vital components. Ericsson posted a nearly
disruptions ranging from geo-political instability $1.7 billion loss for the year, and
in the supply region to more localized factors ultimately had to outsource its cellular
such as strikes at the source of supply. The handset manufacturing business to
following high profile events provide an insight another firm [7].

Globally, billions of dollars have sunk in

supply chain disruptions due to lack of or weak
supply risk mitigation strategies

into severe supply disruption losses due to
supply chain risk:
• In 2004, a strike by 500 workers at an auto
parts supplier forced General Motors
to idle two of its Canadian factories which
employed about 6000 workers, costing
millions of dollars in losses [5].
• About 37 percent of Dow’s North
American production capacity, which
totals 1.35 billion pounds of polypropylene
annually, was affected by Hurricane
Katrina in 2005 [6].
• In March 2000, a Philips manufacturing
facility in Albuquerque, New Mexico,
was destroyed by fire; the facility
As can be gauged from the above
examples, a series of events produce a domino
effect and cause disruptions to supply. There are
several risk categories which have to be analyzed
for potential impact - in isolation, as well as in
combination, with other contributing factors.
Supplier risk is one category that is localized
within the source of supply and can include
supply delays due to strikes, non-performance
of products/components against specifications
and non-compliance to contracts. A second risk
category is related to logistics from the supply
source to destination including factors such
as distribution infrastructure breakdowns,
communication breakdowns with logistics
and supply partners due to hardware or

94
 Profile Prioritize & Implement & Innovate &
Supply Risk Plan Monitor Improve

Exhibit 1: Supply Risk Management Framework Source: Infosys Experience

software issues and cross-border shipment delays various business objectives are working at
due to changes in customs and other statutory cross purposes. In order to be more responsive
regulations. The third, and most unpredictable to market, companies have invested in
risk category is related to macro-factors product diversification and to minimize time
such as sudden geo-political instability or to market they match product delivery rate
natural disasters such as hurricanes or to demand. However, with a view to cut cost,
earthquakes. companies have outsourced business operations,
Supply risk factors are further minimized inventories, and made operations
amplified when considered in the context of cost very lean. The challenge is in maintaining cost
efficiency strategies that have been adopted by efficiency without compromising on schedules,
several manufacturing companies. Strategies quality and market response rates.
such as consolidation of supplier base, lean One way to align both objectives
manufacturing and JIT pose serious threats is to have a consistent and reliable source of
in case of disruption to normal routine. To supply. This essentially means identifying and
add to this, legislation compliance has added resolving supply issues as early as possible.
a complexity factor due to international labor While this is conceivably easier if supply
laws, environmental regulations and volatility sources are very close to the destination,
in global trading. the complexity increases in a global supply
Clearly, global sourcing decisions chain. Global corporates need to develop and
cannot be made purely on a total cost basis alone. follow an all-encompassing and holistic risk
Risk analysis needs to be an inherent part of any management model - one that looks at all the
global supply scenario analysis. uncertainties and their degrees of influence on
the various segments of the global supply chain.
FRAMEWORK FOR ENTERPRISE SUPPLY
RISK MANAGEMENT The following exhibit provides a proposed
Supply risk management is a systematic process framework for supply risk management:
of managing unwanted events or unwanted
change in the supply chain and developing more Profile Supply Risk
predictability in supply. In order to understand and mitigate risks,
The challenge in supply risk corporates need to generate supplier risk
management is that the strategies taken for profiles. An intuitive and effective way

95
 Risk Product (RP)

Severity Impact

Likelihood Negligible Marginal Serious Critical Catastrophic

1x5= 2x5= 3x5= 4x5= 5x5=

Highly Probable
5 10 15 20 25
1x4= 2x4= 3x4= 4x4= 5x4=
Probable
4 8 12 16 20
1x3= 2x3= 3x3= 4x3= 5x3=
Occasional
3 6 9 12 15
1x2= 2x2= 3x2= 4x2= 5x2=
Remote
2 4 6 8 10
1x1= 2x1= 3x1= 4x1= 5x1=
Improbable
1 2 3 4 5

Figure 1: Sample Profiling of Risk Source: Infosys Research

is to review the processes within a given

High
supply chain and segregate the risk based on
likelihood of occurrence and severity of impact.
The steps to profile supply risk are outlined
below [8]: Plan Partner

i) Prepare a process map of the overall Final

Imapact
supply chain and sub processes with the
Leverage Manage
help of key stakeholders, critical supply
chain system, process and production
activities;
ii) Determine likelihood of unwanted
variation (risk events); Low
High
iii) Identify which sources of variation Supply Criticality
represents higher severity;
Figure 2: Risk-Response Matrix
iv) Determine risk product profile.
Source: Infosys Experience

96
 RP Risk Action Control

R1 – Sole source vendor A1 – Determine vendor C1 – Independent

in tornado alley recoverability validation of recovery
20
A2 – Find second source capability. Reevaluate

R2 – Non-ISO certified A1 – Re-negotiate C1 – Require proof of

vendors contracts requiring ISO certification by 3rd party
16
certification bi-annually

R3 – Pending legislation A1 – Require vendor/ C1 – Monitor delivery

15 adds 2 days on-dock shipper to meet earlier times and on-dock wait
time for key components shipping schedule

R4 – Critical component A1 – Increase insurance to C1 – Monitor insurance payout

lost shipments offset lost revenue vs lost revenue
12
A2 – Work with vendor/shipper C2 – Implement vendor controls
processes to improve tracking to monitor processes and data

Figure 3: Prioritize Risks – Identify Controls Source: www.QualityPlusEngineering.com

Manufacturers need to repeat the
profiling process regularly in order to have
a realistic picture of risks in the supplier base.
Figure 1 shows a sample profiling of risks based
on likelihood of occurrence and severity of
impact.
Prioritize and Plan
Based on the risk profiles that have been
generated, risk managers can assess, prioritize
and plan the risk response and update them to
the corporate risk database.
Developing a risk response plan can
be challenging and the approach would vary
from supplier to supplier based on criticality of
the supply and the financial impact as shown
in Figure 2. A part with high supply criticality
in this case is defined as one which has high
entire switching costs and/or does not have many
alternative sources of supply.
Risks for suppliers falling under low
material criticality and low financial impact can
be countered by leveraging the organization pull
for better contracts and by demanding a higher
level of risk insurance from the supplier. For
example, if the supply is an MRO component
such as a standard fastener, the onus of managing
the supply and inventory can be passed on to
the supplier.
Risks on low criticality parts with
higher financial impact can be countered
through traditional planning and forecasting
cycles. An example of this would be AC motors
used in various consumer goods where there are
numerous sources of supply but costs are not
necessarily low.

97
 For more critical parts with low financial
impact, risks can be countered by managing
inventory effectively. Creating extra buffer
inventory for these parts can typically insure
against any disruptions. Custom molded rubber
parts used in specialized packaging systems, for
instance, may not be expensive on a unit cost
basis but it may be difficult to retool an alternative
supplier under short notice. This would make it
critical from a supply perspective.
Critical parts with high financial
impact require the closest amount of planning
and monitoring. Establishing strategic supplier
partnerships to collaboratively manage risks is
the most effective approach in such situations.
This scenario would apply, for example, in
the case of industrial electronic manufacturers
who buy specialized processors which are both
expensive and have a very limited source of
alternative supply.
Figure 3 shows sample risks along
with a number representing the risk product
(RP), potential actions and controls. Action
denotes the steps recommended by the risk
experts to respond to the risk and control
denotes the checks and balances placed in the
supply chain to ensure the effectiveness of the
risk response plan.
Implement and Monitor
The success of the risk management program
lies in the effectiveness of execution. Tools
and techniques that can be utilized to aid in
implementation and monitoring are key to
effective execution.
Given a specific approach, a number of
tools and techniques can be utilized depending
on the specific situation. Many of the available
tools are fairly intuitive and are used in some
form or the other by most organizations. Tools
by themselves are just enablers, and need to
98
be applied appropriately in the context of the
specific risk control plan, in order to maximize
their effectiveness. Following are some best-in-
class implementation tools and practices:
Risk management data base -This is a collation of
statistics and data from each project to be stored
in an organizational risk database. Value-added
information such as choices or options that were
available and decisions taken that were taken
to mitigate the risk and their success could be
documented. This historic information is made
available for future projects to help make right
decisions at the right time.
Global supplier directory: This is a large
compendium or reference document of latest
information spanning suppliers of components,
features, costs, performance and contact
information across different geographies. This is
useful in supplier selection and audit phases.
Designing redundant systems: While creating
the supply chain master plan, designing
redundant systems will allow flexibility of
operations by allowing for alternate supply
sources [9]. Making systems lean is a risk due to
the fact that a local disruption or event can cause
the entire supply chain to be at risk. In case of
labor strike, acts of God or political unrest this
redundancy helps in re-configuring the supply
chain in response by taking an alternate supplier.
This approach would be recommended in case
of highly critical components where the impact
of any supply failure would be catastrophic.
Susceptibility of network calculator is a tool to
determine the impact of a natural or man-made
disaster on the individual locations of a supply
chain and also the combination of them. This
helps in gauging extent of damage to take risk
 Categories Rating Comment
Business competency
Technical competency
Competitive Price
Competitive Service
Schedule
Supplier Score Card Quality
Scope
5 Excellent Cost
Communications
4 Above Average
Issue/Risk
3 Average
Collaboration
2 Below Average
Disaster recovery & BCP
1 Poor Overall

Exhibit 2: Supplier Scorecard Source: Infosys Research

avoidance or transfer options or steps to limit
the risk exposure. For e.g., to open a plant in an
earth-quake prone area, one may need to know
the amount of insurance cover to be taken or the
size of the plant to be built given the likelihood
of the inherent risk. It will also aid in planning
the contingency fund that the management can
budget for, in case of an emergency occurrence.
Supplier quality audits are performed by the
procurement managers to see if quality and
performance requirements will be met. During
the planning phase, the schedule for audit,
purpose of audit, persons to be audited, samples
to be inspected could be decided. The timing of
the audit should be such that there is a possibility
of corrective and preventive action being taken if
audits bring out high risks.
Pareto analysis to weed out ineffective global
players. 80 % of problems may be caused by 20 %
of suppliers. The analysis would help in isolating
these trouble points, providing opportunities to
improve by giving time. If these suppliers do
not perform to the expected level then they can
be removed from the list of suppliers. In cases
where 80% of high quality products come from
20% suppliers, rewards to such suppliers could
help keeping their motivation levels high.
Supplier scorecard is a tool to rate performance
of suppliers. Additionally supply chain would
be at risk if a supplier goes bankrupt or has cash
flow problems. So, it would help to monitor the
financial health of the supplier since this could
result in potential supply risk. The parameters
that indicate the quality of service and products
could be measured as in the sample indicated
in Exhibit B and corrective action can be taken
as needed.
Once the audit results are published,
certain corrective action needs to be taken. Based
on the risks that occurred and the supplier’s
performance, manufacturers could re-negotiate
contracts to include clauses to ensure business
continuity planning and penalties for non-
conformance or poor performance.
Software tools – Use of Information
Technology in GSC risk management is
three tiered:

99
 OEM
Delivery
Rate from
Material Production Shipment
Powertrain
Usage Rate Rate Rate

Parts Inventory Work-in-Progress Finished Product Produc-in-Transit

Manufacturing
Delivery
Management
Go To Powertrain
Production
Scheduling

Order Rate to
Powertrain
Material Order
Management Scheduling
Dealer Orders (External to
model)

Figure 4: Screen Shot from GoldSim Supply Chain Model. Source: www.goldsim.com

Tier 1 - Tools for operational efficiency: softwares help diagnose problems, evaluate
These are tools that deal with flow options, optimize operations, and mitigate
of data and information at the operational risk factors. The objective is to interpret
level. Examples are tools to handle inventory the influencers and interrelationships of
management, network design, product the system with respect to time and use it
life cycle management, demand visibility, to predict the future behavior of the system
supply visibility, demand planning, sales and or determine what factors to vary to achieve
operations planning. a desired result state. One such example
is the GoldSim simulation software package
Tier 2 - Analytic tools: These are decision that uses time-dependent simulation to
support tools used for proactive analysis predict the full range of possible futures,
conducted by analyzing and mining analyze results and communicate the
information obtained from operational level. findings to stakeholders and decision-
makers [10]. The screen [Fig. 4] depicts
Tier 3 - Strategic tools: Modeling and the conceptual model for an OEM.
simulation packages would be the prime tools Quantities that are tracked within the
in this category. Dynamic simulation model include: parts inventories, backlog,

100
 work-in-progress, finished product, and
product-in-transit.
The other emerging trend in this
category is predictive analytics – Open Ratings
(a Dun & Bradstreet Subsidiary) mines supplier
data from a variety of sources and runs statistical
analysis which can predict the risk of supplier
failure. It looks beyond financials into factors
such as on time delivery and product quality
to predict the suppliers most likely to have
problems [11].
Innovate and Improve
The above tools and techniques if implemented
effectively and in a structured manner should
be fairly effective in insuring against supply
risks. The next stage in the evolution of supply
risk management goes beyond the first tier to
subsequent tiers in the supply chain and involves
an increased level of collaboration with suppliers.
The key to innovation and improvement
lies in knowledge and experience sharing with
specific suppliers and about those suppliers’
suppliers.
Information tracking and collaboration portals
move beyond immediate suppliers to the next
tiers down the chain. Information workflow
and shared knowledge portals help in reducing
uncertainty. Manufacturers could get visibility
into status of parts and material shipments
and also view legal documentation related
to the supply from the portal. Notification of
service interruptions can be posted to help
indirect suppliers and customers be aware
of the event and also understand the business
impact on them. Supply chain managers from
leading electronics companies confirm that
globalization is affecting everyone in the
electronics industry and that collaboration
across organizational silos and supply chain
101
partners is critical to a successful product
introduction and management.
Forecasting of aggregates - Another risk
reduction example in the supply chain planning
cycle is the use of aggregates or sub-assemblies
instead of raw materials to reduce forecasting
risk. The principle behind this is forecasting
of aggregates is more reliable than forecasting
single products.
Supplier self-assessment - A third example of
collaborative risk reduction is the process of
self assessment by suppliers on their internal
constraints. Suppliers can complete this and
share the information with their immediate next
set of suppliers and customers. This ensures a
minimal level of awareness of performance risks
within segments of the supply chain. If there are
no plans for contingencies, then manufacturers
need to work with the suppliers to come with a
business continuity plan.
Use of SCOR - Supply Chain Operations
Reference Model as a process reference model
is a standard and effective way of sharing
supply chain information among partners [12].
SCOR is an industry standard for management
focus across inter-company supply chains.
SCOR is used to describe measure and evaluate
GSC configurations.
Describe – Standard SCOR process
definitions allow virtually any supply-chain
to be configured.
Measure – Standard SCOR metrics enable
measurement and benchmarking of supply-
chain performance.
Evaluate – Supply-chain configurations may
be evaluated to support continuous
improvement and strategic planning.
Risk appetite baseline – Senior management of
manufacturers need to document their appetite
for risk and risk tolerance limits. This would
outline guidance for downstream operational
managers to take decisions and also bear in mind
the revenue that the management is willing
to risk. This is particularly useful in deciding
on introduction of new suppliers, products
and services and also in sourcing from new
geographic locations.
Disruption drill is the mock disruption of
supply in the supply chain to test the redundancy
of network, process capability and supplier
capability. A supply interruption of a critical
component is mimicked to appraise the supply
chain and also gauge the time taken to recover
from a disaster as well as the cost taken to ensure
business continuity. The disaster recovery and
business continuity plans may be revised based
on the outcome of the simulation. The learning
from the exercise could be shared through the
collaboration portals.
Innovation lies in the understanding
that the success of the manufacturer lies in the
success of the suppliers. Supplier base could be
reviewed to decide on sole-sourcing versus
multi-vendor sourcing strategy. For example, if
a local supplier’s performance is better, but
costs more than a global supplier’s, the supply
chain strategy could be to avoid global
outsourcing of critical components and source
them locally to mitigate risk. The framework
provides an innovative paradigm shift from
cost reduction to supply chain continuity.
COMPONENTS OF THE GLOBAL SUPPLY
RISK MANAGEMENT MODEL
Just like in any enterprise initiative, people,
process and technology components of risk
management need to be clearly identified and
102
considered prior to embarking on a supply
risk management program [13]. Figure 5 below
depicts the components of the Global Supply
Risk Management model.
People - At each node or entity of the supply
chain there are set of key stakeholders including
operational managers, third party logistics
providers, purchasing analysts and risk experts.
The roles and responsibilities of each of these
stakeholders within the risk management
program need to be clearly defined. For example,
does purchasing own the supplier risk profiling
or is it the operational managers? What is the
role of business continuity managers in ongoing
risk management? All these need to be clearly
mapped out upfront to prevent confusion.
Process – The success of a program lies
in the effectiveness of execution – process
standardization is critical especially in large,
global enterprises. Ideally, supply risk
management processes are streamlined across
the entire global supply chain but are agile
enough to adapt to specific supply situations
Tools and technology – Tools and technology
act as enablers of any people and process based
initiative. Many of these have been discussed
under the ‘Implement & Monitor’ section earlier.
Governance is achieved through a central body
that stresses compliance to a set of performance
and quality standards. Metrics for monitoring
supply risk are often very similar to regular
operational metrics (price, quality, delivery).
However, the metrics for measuring the
effectiveness of a risk management program
would be unique. Some metrics include:
• Ratio of actual financial impact to
potential impact of disruption
 Tools Processes

Risk S.
Database Profiling
Demand/ Risk
Supply Visibility Profiling
Simulation & Scorecard
Modeling Site Audits
Trend
Collaboration
Monitoring
Global Supply Forecasting
Data Mining
Risk Management Network
R. Reporting
Model Design
Monitoring Senior Mgmt
Mechanisms Review

People Operations Purchasing Process

Risk Analysts 3PL Managers
Managers Managers Consultants
Governance
Contract Disaster Knowledge Innovation Quality Process
Management Management Management Management Management Management

Figure 5: Components of the Global Supply Risk Source: Infosys Research

Management Model

• Ratio of risk redundancy cost to potential the right balance between operational
impact of disruption efficiency and risk
• Lead time to contain disruption. • Senior management commitment to risk
management
PRE-REQUISITES FOR AN EFFECTIVE • Customize and SCRM strategy specific
IMPLEMENTATION to your organization - change it based on
Changes to organization or operations are the business environment
difficult to implement as they require a change • Analyzing and learning from supply
in the mind-set of the persons affected. The disruptions to increase supply chain
framework provided here attempts to change continuity
several aspects not only in a single organization • High level of trust and confidence in
but spanning across organizations. For a change sharing information and co-operation
of this magnitude to be accepted and embraced between business entities. This can be
by all, it is essential to identify certain key factors. very difficult to achieve if the customer
The critical success factors for the model are: supplier relationship is competitive and
• Continuous and concerted effort in based on price alone. But once they
identifying and managing risks and see their relationship as a long term
ensuring business continuity win-win partnership possibilities for
• Constantly measure and strive to attempt collaboration automatically open up.

103
CONCLUSION 3. American Metals Market Index, www.
Business disruptions in global and complex supply amm.com.
chains have tested the manufacturer’s resilience 4. Harold L. Sirkin et.al., What is
due to global sourcing and lean manufacturing. Globalization doing to your Business?
Manufacturers can benefit from increased Boston Consulting Group , Feb 2004.
transparency and communication, business Available at http://www.bcg.com/
continuity, performance-based contracts and publications/filesWhat_Is_
reduced risk exposure by adopting this model. Globalization_Doing_to_Your_Business_
Manufacturers need to encompass first-tier, Ops_Feb04.pdf.
second-tier suppliers and understand source- 5. Tentative deal reached after strike at parts
to-destination issues in global supply chains. supplier idles pair of GM plants, CBC
There is a need to closely work with suppliers Canada , July 2004.
and vendors and develop business continuity 6. Supply Chains in Katrina’s wake, CFO
plans. The risk management model is a reference Magazine, September 2005.
for transforming the supply risk management 7. Flexibility in the Face of Disaster:
process from a reactive exercise to a proactive Managing the Risk of Supply Chain
practice. A robust, systematic and multi Disruption, Knowledge@Wharton,
enterprise-wide approach to risk management Sept 06, 2006.
is essential to meet business objectives in a 8. www.QualityPlusEngineering.com-
flat world. Early and continuous focus on Applying Risk Management to the Supply
risk management will be critical for operational Chain - WESTEC.RM. Applying RM to SC-
success and predictable continuity of a Stan.2”, Apr 2005 -Stan Smith - Risk
supply chain. Collaboration and increasing Assessment Consultant, Quality Plus
information transparency reflect the collective Engineering, LLC.
resolution of the supply chain industry to resolve 9. David Simchi-Levi et. al., Designing
its problems. The GSC Risk Management model and Managing the Supply Chain, McGraw
offers a pro-active and pragmatic approach to Hill, International Edition, 2003.
mitigate risks in Global Supply Chains. 10. Dynamic simulation and supply chain
management, GoldSim Technology Group
REFERENCES Report, April 2005, www.goldsim.com.
1. David Huether, The State of 11. www.industrynews.com.
Manufacturing: A review of the major 12. www.ascet.com
indicators, National Association of 13. V S Srividhya and Ananth Subramanian,
Manufacturers, March 2007. 360 Degree RISK Management
2. Erin Lett and Judith Barrister, Labor Model – A new model to Rate, Mitigate
costs of manufacturing employees in and Exploit Opportunities, QAI PML
China : An update to 2003-2004, 2004. Conference Proceedings, 2007.

104
 SETLabs Briefings
VOL 5 NO 3
Jul-Sept 2007
A Supply Chain Collaborative
Solution using XQuery
By Bharani Shivaraja Govindasamy
This new W3C specification promises a solution
for data aggregation challenges in CPFR
S
upply chains encompass the companies and
the business activities needed to design,
make, deliver, and use a product or a service.
Businesses depend on their supply chains to
provide them with what they need to survive
and thrive. Every business fits into one or
more supply chains and has a role to play in
each of them. The pace of change and market
uncertainties have made it increasingly important
for companies to collaborate and hence be aware
of the supply chains they participate in and to
understand their roles. Collaborative supply
chains are becoming complex as the number of
business partners increases while information
needs also increase [1].
Several open standards are defined
for supply chain collaboration. Collaborative
Planning, Forecasting and Replenishment
(CPFR) is one business model developedby
Voluntary Inter-industry Commerce Standards
(VICS). CPFR defines a set of business processes
for enterprises to collaborate in planning,
forecasting and replenishment activities to arrive
105
at a common order. To achieve this, enterprises
share their forecast information, promotion
information and point of sale (POS) information,
which is used to identify variations in forecasts
for monitoring the supply chain and to
calculate key performance indicators (KPI) for
measuring the supply chain performance [2].
Internet-enabled enterprises are
beginning to use XML for information sharing
as XML offers a structured format for transporting
data over the internet [3]. Enterprises that
collaborate with business partners require
information from disparate data sources to
be integrated [4]. Existing applications that
implement CPFR, however, are difficult to
maintain and are not scalable. An application
that can aggregate data from diverse sources
and present it as actionable information helps in
making quick and intelligent decisions [5].
This paper proposes an application
architecture using XQuery to aggregate data
from disparate data sources. For this purpose,
several CPFR scenarios are identified in whcih
two retailers share their sales forecast, promotion
forecast, order forecast, POS and promotion
information with the manufacturer. XQuery
programs are written to execute the scenarios and
obtain the aggregated data containing forecast
variations as exceptions and KPIs. A dashboard
application is developed using Visual FoxPro
which provides rich graphical interface to view
the aggregated data. This enables executives to
manage their supply chains better.
PRECURSOR
Supply chain is a series of links and shared
processes between suppliers and customers.
These links and processes involve all activities
from acquisition of raw materials to delivery
of finished goods to the end consumer. Raw
materials enter a manufacturing organization via
a supply system to be transformed into finished
goods. The finished goods are then supplied
to consumers through a distribution system.
Generally, several companies are linked together
in this process, each adding value to the product
as it moves through the supply chain. Since the
value addition for a product does not rely on the
manufacturer alone, the concept of supply chain
management assumes significance [6].
Traditional SCM techniques assumed
that the information is available for an immediate
supplier and customer in a business. This meant
a linear supply chain where information sharing
was only between a stakeholder’s immediate
participants in the supply chain. This resulted in
a bull whip effect — that refers to the increase
in demand variability as one moves up the
supply chain away from the retail customer.
This meant that small changes in consumer
demands resulted in large variations in
orders placed upstream. Studies on managing
the bull whip effect pointed to the need
for a collaborative supply chain where all
106
the participating business partners share
information and make it available to all the
other business partners [7]. This would mean
that information visibility across the supply
chain enables companies to notice the variations
in demand that arise as a result of logistics
issues, plant shut down and promotions in
real-time and hence plan for managing the
variations in advance.
Information exchanges assist in
transporting data across the supply chain to
be used by all business partners. Traditionally,
Electronic Data Interchange (EDI) was used
to enable businesses to build information
systems (IS) and share data in a format agreed
upon by all the partners. The information
exchange was through networks such as WAN.
The disadvantage of EDI was that it was too
expensive to implement and the businesses
were forced to build separate systems for each
of their partners. Also, the business software
applications across the enterprises used disparate
software components making integration of IS a
formidable task [8].
The need therefore was for data and
communication standards for information
exchange [8]. This paved the way for web
services that advocate interoperability between
software systems and offer a standard message
format and communication medium [9].
Web services are software programs
that can interact with the other software
programs via the internet using open standards.
They are loosely coupled which means
integrating applications that are built using
disparate components is easy and hence data
and application integration can be effectively
managed [9]. Web services use extensible
mark-up language or XML as the standard for
sharing data. XML was developed as a means to
describe data and has now evolved into a data
standard as a result of its ability to support data
transportation through the internet [10].
As companies choose to implement web
services to enable supply chain visibility, need
now is to integrate information that is available
from the web services at the business partner
end and the enterprise information available
in the databases at the stake holder’s end
in real-time or near real-time to facilitate decision
making. Enterprise Information Integration (EII)
has evolved into a business strategy and deals
with this task of integrating disparate data
sources such as XML from web services and
relational data from databases. EII focuses on
the integration of data from multiple systems
into a unified, consistent and accurate
representation of data that can be viewed and
manipulated as the need may be. The type
of data has to be real-time which means that
the information from different data sources have
to be processed and made available as and when
it is needed [11].
Technologies such as JDOM , which
are built on programming languages like
Java, are capable of manipulating data from
different data sources [10]. Software programs
can be written to access the data from
different sources using Application Programming
Interfaces (API), each written for the purpose of
accessing and processing data from a specific
source.
The other type of languages that
perform the task of accessing and processing
XML are called query languages. Unlike
programming languages that are descriptive
(where we specify how we want it done), query
languages are declarative meaning one just
has to specify what is wanted [10]. World
Wide Web Consortium (W3C), a non-profit
organization for developing open standards for
communication over the Web, has developed
107
a declarative query language called XQuery.
XQuery holds the promise of efficiently
processing data by joining, and sorting
information from disparate sources [12].
Although XQuery is not yet a specification,
its working draft version enables one to explore
the possibility of enterprise information
integration in detail. As information integration
is important for supply chain collaboration,
we will focus our study on evaluating XQuery
for supply chain collaboration.
COLLABORATIVE PLANNING,
FORECASTING AND REPLENISHMENT
Supply Chain Collaboration
VICS defines a standard for combining the task
of planning, forecasting and replenishment for
increased efficiency in a supply chain [Fig. 1].
Their CPFR model formalizes the processes
between two trading partners who agree upon a
joint plan and forecast, monitor success through
replenishment, and recognize and respond to
any exceptions. CPFR builds upon Efficient
Consumer Response (ECR) principles including
Vendor-Managed Inventory (VMI), Jointly-
Managed Inventory (JMI) and Continuous
Replenishment (CRP). These process standards
involve exchange of information between
enterprises [2].
CPFR can be defined as a business
process wherein trading partners use technology
and a standard set of business processes for
internet-based collaboration on forecasts and
plans for replenishing product.
For the purpose of this study, we use
CPFR as the reference to define business
scenarios where trading partners — retailers,
warehouses and manufacturers — have
visibility into one another’s critical demand,
order forecasts and promotional forecasts using
web services [2]. CPFR models address the
 1. Front-End Agreement
Collaborative Planning
2. Joint Business Plan
3. Create Sales Forecast
4. Identify exceptions
5. Resolve exceptions
Buyer
Figure 1: The 9-Step CPFR Process Model
supply chain inefficiencies caused due to lack
of information sharing among trading partners.
The need for CPFR models is more pronounced
as information must be available quickly to
all trading partners to take timely action and
meet unexpected events. For example, a spike
in demand will register in time for everyone in
the supply chain to effectively adjust and ensure
proper inventory levels on store shelves. Costs
for both trading partners are greatly reduced
because problems can be anticipated and
corrected proactively. CPFR provides a logical
solution to resolving historical supply chain
inefficiencies. The internet has provided a means
for instantaneous exchange of information
between trading partners. Its use, in conjunction
with the industry-defined CPFR processes, has
enabled a systematic method to reduce supply
and demand chain inconsistencies [2].
108
Seller
Collaborative Forecasting
6. Create Order Forecast Collaborative
7. Identify exceptions Replenishment
8. Resolve exceptions
9. Generate Order
Source: http://www.vics.org
CPFR defines a 9-step process model to
arrive at a common order [Fig. 1]. This involves
steps for collaborative planning, forecasting and
replenishment.
XQuery
The explosive growth of XML-based standards
bears testimony to XML’s interest to many
different technical communities. As the volume
of information stored in XML grows, it becomes
correspondingly more important to be able
to access information in XML documents
efficiently and effectively. To do that, we
need an expressive query language so we can
specify the information we want to retrieve
from or update to an XML data source. XQuery
intends to be that language [10]. It is designed
to query collections of XML data— not just
XML files, but anything that can appear as
XML, including relational databases. XQuery $actualorder.Orders_2/weeknum)
provides the mechanism to efficiently and easily return
extract information from Native XML Databases <kpi>
(NXD) as well as from relational data. With <itemid>{ xf:data($orderforecast.forecast_1/
XQuery, we can view RDBMS tables XML itemid) }</itemid>
data source. XQuery offers the possibility of a <weeknum>{ xf:data($orderforecast.forecast_1/
single query that combines an incoming purchase weeknum) }</weeknum>
order in native XML format, an archive of catalog <forecast>{ xf:data($orderforecast.forecast_1/
data also in native XML format, and an inventory quantity) }</forecast>
system held in a relational database [6]. <actual>{ xf:data($actualorder.Orders_2/
Consider the example where XQuery quantity) }</actual>
code can be put to use in a CPFR environment. <accuracy>{ cast as xs:decimal(cast as xs:
A manufacturer wants to calculate the order integer($v_8)) }</accuracy>
forecast accuracy using the data available in </kpi>
relational databases and flat files. An XQuery }</results>
code as shown below can be used to aggregate In the above XQuery code, the
the data and calculate the order forecast accuracy, <accuracy> tag in the output XML will have the
a KPI in a CPFR environment. information of the order forecast accuracy. We
The formula used to calculate order can now see that XQuery can be used in any
forecast accuracy is as follows: Order forecast application where there is a need to aggregate or
accuracy in % = [1-{(order forecast quantity-actual process XML data. Most of its uses fall broadly
order quantity)/order forecast quantity)}]*100 under the following areas:
We will use the XQuery operators 1. Querying XML data sources: A digital
such as -, div, * and cast_as_xs:float and cast_ content marketplace where XML content
as_cs:integer to calculate the key performance is managed in the case of single source
indicator. multiple presentation scenarios.
<results>{ 2. Integrating data: Enterprises that use data
For $orderforecast.forecast_1 in document from a mix of data sources will use
(“orderforecast”)/orderforecast/forecast XQuery for a data integration solution.
for $actualorder.Orders_2 in 3. Integrating applications: Software
document(“actualorder”)/OrderInfo/Orders applications talk in different l a n g u a g e s
let $v_3 := $actualorder.Orders_2/quantity and need to be integrated. XQuery
- $orderforecast.forecast_1/quantity provides the transformation capabilities
let $cast_as_xs:float_4 := cast as xs:float($v_3) where a source document can be searched
let $cast_as_xs:float2_5 := cast as xs: and a result document structure
float($actualorder.Orders_2/quantity) of choice can be built [10].
let $div_6 := $cast_as_xs:float_4 div $cast_as_xs:
float2_5 CPFR DASHBOARD APPLICATION
let $v2_7 := 1 - $div_6 In a typical CPFR application, we assume six
let $v_8 := $v2_7 * 100 different scenarios for a supply chain to evaluate
where ($orderforecast.forecast_1/weeknum eq the concept of data integration via XQuery.

109
 XML Liquid Data XQuery Engine
Retailer 1
Web Service
XML
XML
Retailer 2
Web Service
Database
Retailer End
Figure 2: Application Architecture Used to Implement
CPFR Scenarios
These scenarios use forecast and point-of-
sale information to determine KPIs
exceptions. Three of the scenarios consolidate
data into KPIs to assess the quality of
business processes in place, while the other
scenarios are that of generating exceptions. The
exception scenarios use the sales/order forecast
data for the next eight weeks while the KPI
scenarios work with eight preceding weeks of data.
Three distinct types of data sources
have been used to represent the diversity
of data sources in a CPFR supply chain
environment. Information from the retailers is
assumed to be available via their web services
applications and/or equivalent XML data.
Manufacturer’s inventory and forecast data
are available in SQL databases while actual
orders and available to promise (ATP)
information is assumed to be available in CSV/
Excel files. Each data source is automatically
converted to or made available as XML data.
XQuery programs are written to perform the
XML
Web Service Supply Chain Dashboard
(Visual FoxPro)
XML
Excel/CSV files
Manufacturer End
Source: Infosys Research
data aggregation on the XML data. Liquid Data
and by BEA systems was used as the XML engine to
convert the data sources to XML and to process
the XQuery programs [13]. The evaluation
version of Liquid Data software provided the
functionality to configure the data sources to
develop XQuery programs, and execute them
in this study. Table 1 summarizes the data
sources and the results achieved. The results of
data aggregation by XQuery engine are made
available via six different web services. Figure
2 summarizes the data flow within each of the
six scenarios. All results are stored in an SQL
database and/or presented on a dashboard for
appropriate business decision making.
CONCLUSION
Data for supply chain collaboration resides
in disparate data sources. An application
architecture using XQuery is suggested for
implementing CPFR scenarios, where functions
and expressions of XQuery are used to aggregate
110
 # Data Sources Used
1 1) An XML file which holds the POS
information for eight Weekly periods is
considered as an equivalent to a web
service response from each retailer.
2) A SQL server database of the
Manufacturer contains the Sales
forecast formation
2 1) An XML file that holds the information
about the sales forecast for the next
eight weekly periods is considered
as an equivalent to a web service
response from each retailer.
2) A SQL server database of the
Manufacturer contains the Sales
forecast information
3 1) An XML file which holds the POS
information for eight weekly periods is
considered as an equivalent to a web
service response from each retailer.
2) An XML file which holds the on-hand
inventory information for eight weekly
periods is considered as an equivalent to
a web service response from each
retailer.
3) A SQL server database of the
Manufacturer contains the promotion
information for the previous eight weeks
4 1) An XML file that holds the information
about the promotions for the next eight
weekly periods is considered as an
equivalent to a web service response
from each retailer.
5 1) A CSV file contains the information
about the orders for each item for the
previous eight weekly periods.
2) A SQL server database of the
Manufacturer contains the Sales
forecast information
6 1) An XML file that holds the information
about the order forecast for the next
eight weekly periods is considered as
an equivalent to a web service
response from each retailer.
2) A SQL server database of the
Manufacturer contains the Order
forecast information
Table 1: CPFR Scenarios Used for Data Aggregation
Processing Done
Sales forecast accuracy in % =
[1-{(sales forecast quantity-
POS quantity)/(sales forecast
quantity)}]*100
Compare the sales information
available in the XML file for
each retailer with the data in the
manufacturer’s database for each
retailer. If the sales forecast variation
is more than 10%, the system flags
the output for that weekly period to
contain an exception.
Out of Stock (OOS) frequency in % =
[number of periods during promotions
when the item went OOS / number
of periods when promotions were
held]*100
Promotion forecast accuracy in %
= [1-{(promotion forecast quantity-
actual POS quantity during the
promotion period)/(promotion forecast
quantity)}]*100
1) An XML file that holds the
information about the promotions
for the next eight weekly periods
is considered as an equivalent to
a web service response from each
retailer.
Order forecast accuracy in % =
[1-{(order forecast quantity-actual
order quantity)/(order forecast
quantity))}]*100
Compare the order forecast information
available in the XML file for each
retailer with the data in the
manufacturer’s database for each
retailer. If the order forecast variation
is more than 10%, the system flags the
output for that weekly period to
contain an exception.
Source: http://www.vics.org
111
Results Achieved
An XML file containing the
sales forecast accuracy KPI
for each retailer for eight
weekly periods.
An XML file containing
the differences in sales
forecasts between retailers
and manufacturer and the
exception flag if present.
An XML file containing the
KPIs, OOS Frequency and
Promotion forecast accuracy
for eight weekly periods.
An XML file containing the
promotion conflict exception
flag for the eight weekly
periods.
An XML file containing the
order forecast accuracy
KPI for the previous eight
weekly periods.
An XML file containing
the differences in order
forecasts between retailers
and manufacturer and the
exception flag if present.
data from disparate data sources in real time. 6. John Babb, Supply Chain Management:
CPFR scenarios use data sources as varied as An Executive reference paper, , Clarkston
web services, relational databases and Excel Consulting, 2000. Available at http://
sheets. XQuery programs can aggregate this data www.clarkstonconsulting.com/White
and identify exceptions and KPIs. They are also Paper/SupplyChainMgmt Paper.pdf.
easy to maintain and scalable. The aggregated data 7. Hau Lee et. al., The Bullwhip Effect
can be presented graphically using a dashboard in Supply Chains, Sloan Management
application. These dashboards help executives Review. 38 (Spring), 93-102, 1997.
to monitor and measure the supply chain 8. Overview of EDI benefits and draw
performance. backs, Covalent Works, Covalent, 2004.
Available at http://www.covalent
REFERENCES works.com/what-is-edi.asp.
1. Mentzer et. al., Defining Supply Chain 9. Get a grounding on the basic
Management, Journal of Business concepts, IBM Corporation, IBM, 2003.
Logistics,Vol. 22, No. 2, p. 18, 2001. Available at http://www-106.ibm.com/
2. Collaborative Planning, Forecasting and developerworks/library/ws-star
Replenishment: Version 2.0, Voluntary there.html.
Industry Commerce Standards, 2002. 10. Michael Brundage, XQuery: The XML
Available at http://www.vics.org/ Query Language, Addison Wesley, 2004.
committees/cpfr/voluntary_v2/CPFR_ 11. John Taylor, Thoughts from the
Tabs_061802.pdf. Integration Consortium: Enterprise
3. Ravi Trivedi, Web services Understanding Information Integration: A New
XML and XML Schema, Jupitermedia Definition, 2004. Available at http://
Corporation, 2003 Available at http: www.dmreview.com/article_sub.
//www.developer.com/services/article. cfm?articleId=1009669.
php/2195981. 12. XQuery tutorial: Building XQuery based
4. What is Enterprise Information aggregation and reporting applications,
Integration?, Advisors, Inc., The Gilbane Data Direct Technologies, 2005. Available
Report, 2004. Available at http://www. at http://www.stylusstudio.com/
gilbane.com/artpdf/GR12.6.pdf. xquery_tutorial.html.
5. David Anderson and Hau Lee, The 13. BEA Liquid Data for WebLogic 8.1
Internet enabled supply chain: From the Documentation – Concepts Guide,
first click to the last mile, 2000. Available 2005. Available at http://edocs.bea.
at http://www.ascet.com/documents. com/liquiddata/docs81/pdf/prodover.
asp?d_id=199. pdf.

112
 SETLabs Briefings
VOL 5 NO 3
Jul-Sept 2007

SOA Based Integration for

Creation of Collaborative Supply
Chain in Automotive Domain
By Krishnendu Kunti, Terance Dias and Ashwini Jeksani

Enable collaboration among supply chain

partners using SOA-based integration platform

A utomotive industry has one of the most

complex supply chains primarily because
of the complex nature of the end product,
number of supply chain partners involved and
the geographical dispersion of supply chain
partners. A typical vehicle requires somewhere
around 5,000 to 7,000 finished parts; these
finished parts in turn consume nearly 130,000
subparts [1]. These parts are either manufactured
in-house or sourced from multiple supply chain
partners across the globe resulting in complex
supply chains. Supply chain in automotive
industry has matured over time and is still in a
flux, driven by business needs.
We discuss the nature of supply chain
in automotive industry, the present business
drivers and resulting demands on supply chain
integration. We also propose on how a SOA
based integration platform is best suited to
address these issues. Subsequently, we discuss
the required functionalities to be supported
by the SOA based integration platform and
highlight the major pain points addressed by
such a platform.
EVOLUTION OF SUPPLY CHAIN
AUTOMOTIVE INDUSTRY
During the early 20th century the modern
automotive industry came into existence. One of
the events that brought automobiles within the
reach of general consumers was the introduction
of Ford T model [2]. The T Model was based
on mass production of single base model with
zero customization resulting in huge economies
of scale. Every component required in Ford T
model was manufactured in-house resulting in
a simple supply chain. However this mode of
manufacturing resulted in huge amount of safety
stock for individual components. Also it did not
allow desired customization to the base model
and did not outsource better engineering and
manufacturing skills from other organizations.

113
During the early 1970’s Toyota introduced an
alternative manufacturing model --- a model
that brought together advantages of economies
of scale and customization as per customers’
demands. Toyota’s manufacturing model
was based on principles of just-in-time (JIT)
manufacturing and keiretsu [3]. JIT is based on
pull system that provides information on exactly
what components and sub-assemblies need to be
produced by means of paper based cards known
as kanbans [4]. Keiretsu relies on collaboration
among multiple business stakeholders towards
achieving common business goals. When applied
to automotive supply chain this results in
outsourcing of some functions to other business
Huge synergies can be gained by different
partners in an intricately woven collaborative
supply chain scenario
entities that are better equipped to execute the
function. For instance, an automotive design-
house could design multiple variants of the
same base model faster than an in-house design
team. Extending JIT to supply chain partners
resulted in even better inventory control and
smother response to variations in demand. Using
the Japanese manufacturing system, companies
were able to meet varied customer demands,
with lesser inventory in a shorter period of
time. In recent years apart from reasons of
technical expertise manufacturing has also been
outsourced to other business entities because
of price differentials arising out of location
advantages.
114
Automotive majors have realized that
in order to survive they should strive to get the
best of technology at the lowest of cost and in
the fastest time to market possible. In order to
achieve the target they need to collaborate with
multiple supply chain partners across the globe.
EMERGENCE OF COLLABORATIVE
SUPPLY CHAINS
In a collaborative supply chain scenario,
supply chains of stake holders are intricately
interwoven. Supply chain partners not only
collaborate towards exchanging data for final
goods delivered (e.g., sending information on the
items to be procured) but also share information
leading to visibility across each other’s supply
chains. Virtually collaboration can happen at
any level of business, for instance, OEM getting
discounts for a medium scale supplier or dealers
deciding on promotion in conjunction with OEM
to push a particular make of a vehicle. The final
realization is that business is driven by efficiency
of the whole supply chain rather than efficiency
of a single organization. Any gains made by
one supply chain partner will be passed on to
other members of the supply chain and vice
versa. Essentially participants of a collaborative
supply chain can be seen as a single distributed
business entity that collaborates at multiple
levels of business requirements. The success
towards creation of collaborative supply chain medium size vendors are reluctant to accept such
depends on the ability to share information technology. Also, EDI is not sufficient to capture
across multiple supply chain partners with the different kind of data interchanges required for
least cost and effort. Given the number of supply creation of collaborative supply chain.
chain partners in a collaborative supply chain Requirements to support different
and the fact that often a single entity takes part in data format across multiple applications
more than one supply chain (e.g., single supplier that communicate using different protocols
supplying to more than one OEM) it becomes and are deployed in disparate systems poses
extremely difficult to meet the integration considerable challenge in integration of
requirements in such scenarios. collaborative supply chain .
• Supporting multiple data formats (e.g.,
IT LANDSCAPE SUPPORTING SUPPLY EDI, flat file, any business specific
CHAIN MANAGEMENT defined schema like STAR etc) both
Typically supply chain applications were either within enterprise and external to
home grown or procured from supply chain enterprise (supply chain partners)
vendors. These applications addressed specific • Exposing business interfaces locked in
supply chain functions such as demand planning, supply chain applications as services
factory planning, production scheduling, using standards based syntax (e.g.,
transportation management etc. Within an WSDL)
enterprise if a supply chain suite is procured • Supporting multiple types of end points
from a single vendor, the vendor, more often e.g., JMS, HTTP, .NET, EJB, RMI,
than not, provides an integration application to databases etc
coordinate among its own suite of applications • Support for widely adopted standards
e.g., i2 operational data store [5], or uses EAI for business document interchange e.g.,
tools for integration. However, if within a supply RFC4130 for MIME-based HTTP
chain, applications are procured from multiple EDIINT (aka. AS2) transfers (use for
vendors then there does not exist any standard EDI) and STAR specification for
means of integration between these applications Business Object Documents (XML)
and in such a scenario integration is done by transfer using ebXML and web services
batch mode data extracts or using EAI tools. • Defining common business schema,
Integration based on batch mode data extracts that maps to different application
induces latencies inherent to batch processes specific schemas and required
and creates point to point connections which are transformations
hard to maintain. While EAI tools might provide • Providing means to create process from
real time integration they create vendor lock-in services using workflow languages like
and are often expensive to implement. The other BPEL etc
issue faced in such scenarios is standardization • Manage authentication and
of exchanged data format. EDI [6], an existing authorization across multiple business
approach for data interchange in the automotive domains
industry, is often expensive both in terms of cost • Integration of supply chains with least
and effort to implement and hence small and cost and minimal effort.

115
Often organizations handle integration of supply based format (e.g., Xquery, OQL) and
chain in piecemeal manner using custom code exposing them to standards based interface
(e.g., ABAP for data extraction form SAP ERP definition language (e.g., WSDL) or using
to be used in planning applications) resulting API. Data services layer also provides value
in point to point connections that cannot be added features like caching, profile based data
reused by other applications requiring the filtering and meta data management. Output
same data. from data access services can be translated
and transmitted over required protocol to
SOA BASED PLATFORM FOR SCM destination systems using SOA based integration
COLLABORATION backbone.
A SOA based platform to manage supply chain
integration should handle integration needs at Application level services: Application level
multiple levels of enterprise IT viz., data access services are provisioned from application
layer, application layer, integration layer and API’s. However application level API can only
process layer. be accessed on a particular platform. At this

A highly distributed architectural platform like SOA should

be able to address supply chain integration at all levels of
enterprise IT to achieve maximum SC collaboration

Data services layer: In enterprise supply layer a SOA based integration platform provides
chain application integration, integrators tools for rapid generation of services from
are often faced with a scenario where in the platform dependent API (JAVA, .NET, Legacy
absence of API for data access, point to point etc). Tools enable faster migration of application
custom code needs to be written for data interfaces to SOA and allow incorporation of
movement. For instance, if data needs to be value added features like caching and security.
moved from one or more manufacturing These tools can also be used to auto generate
applications to one or more destination service interfaces (e.g., web services artifacts),
planning applications requiring different provide mapping between service schema
formats, writing point to point data and business data definition schema,
extraction code can be time consuming and provide value added features like caching,
cost intensive and extremely hard to maintain authentication, etc., and finally deploying
in the long run. In such scenario shared generated artifacts. However it is important to
data services[7] platform can be used for note that an application level API can be directly
defining distributed queries in a standards consumed at integration layer as well. The

116
 Process Layer
Definition of coarse grain business function services by creating work-
flows from services in data access, application and integration

Integration Layer

Meta Data Repository

Service Ontology and relations
Standards based handling of integration, routing, content-based routing, multiple
invocation models, distributed security handling and value added features

Application Layer
API for exposing existing applications as services, support for multiple types of
endpoints, protocols, data translation, value added features

Data Services Layer

Standards based distributed data access definition, Meta data management, stand-
ards based interfaces for data query, value added features

Figure 1: SOA Based Integration Platform for Supply Source: Infosys Research
Chain Integration

only differentiator being, if an application
interface on itself qualifies to be exposed as
reusable component rather than just a part of
flow towards creation of reusable service, then
it is better to expose the application interface as
service.
Integration level services: Services at integration
layer create coarser grained business functions
from services at data access layer, application
layer and native application API. Integration
layer is used for transport bridging, content based
routing, translation, provisioning of distributed
security [8], supporting different invocation
and communication model. For example, an
engineering change might be required to be
propagated to planning and inventory systems
__ these systems might exist across multiple
domains and accept data in different formats.
ESB (Enterprise Service Bus) can be used to
handle such scenarios with minimal effort [9].
Finally an integration layer acts as a gateway for
external services and provides / receives data
from external service in their supported formats
over required transports.
Process level services: These services are
coarse grained and are similar to integration
layer services. They can be implemented at
integration layer or using workflow language
based orchestration layer. These services usually
represent business document interchange
interface that internally drive a number of steps
in a process. For instance, submission of demand

117
order function can be exposed as process interface
where a submitted demand is processed by a
set of services at integration, application and
data access layer using production planning,
scheduling, manufacturing, purchasing and
finance functions.
Meta data repository: In enterprises, IT systems
often represent a single business
differently based on different ontological
classifications. For instance, a unique part
identifier might be referred to as partid in one
system and partno in another system. In SOA
individual services. Hence it is important to
manage non functional requirements such
as business activity monitoring, distributed
security, service look up etc.
A BUSINESS USE CASE FOR SOA BASED
COLLABORATION IN SCM
Let us consider a business scenario that involves
entity introduction of a new vehicle for a particular
segment of buyers viz., young professionals,
elderly, householders etc. The process typically
starts with a market survey across multiple
geographies carried out by the OEM along with

Maintenance of service ontology in a metadata repository

can help negotiate confusion related to multiple naming
of a single entity

where a process is created from a number of other stakeholders like dealers, advertisement
services, managing disparities of data format agencies, etc. The outcome is refined into broad
pose a major challenge. This issue can be level features and pricing. The concept is then
addressed by maintaining service ontology communicated to styling for creation of 3D, 2D
and semantic relations in a metadata and clay model. Once there is mutual agreement
repository. These ontology and semantic on the looks and features by stakeholders __
relations can be used at multiple levels of like engineering, marketing, purchase, senior
SOA. For example, service bus can use the management , the concept is said to be finalized.
semantic relations from provisioning of A new vehicle is either designed from scratch
translation or shared data services platform or created by reusing existing platform and
can use these semantic relations for part components with some modifications. The
lookup where a part identifier field is known by design and engineering process necessitates
different names in different systems. availability of systems that allow creation
SOA is loosely coupled and highly and sharing of documents, CAD integration,
distributed architectural paradigm. In SOA a versioning of product design information, work
process might span across multiple user domains flow management, role-based access and API for
and finally success of a process depends on sharing/ accessing data with other applications.

118
 Java/xml/ Europe Manufacturer
web Asia Manufacturer
Design Systems Design Systems
service
Flat file over
Java/xml/
FTP
web
service
Asia
XML format 1/

Service bus at integration layer

Manufacturer
over
JMS prototyping
system
Service bus at integration layer

Flat file over

JMS

Asia Manufacturer Asia Manufacturer Preferred Supplier

STEP design systems
manufacturing system purchase system

XML format 2/
over
FTP Data services Supplier
xml xml purchase systems
over http over http

Asia Manufacturer
planning system

Service ontology,
relations and translations

Figure 2: Typical Integration Scenario in Supply Chain Source: Infosys Research

Collaboration

Sharing information with other systems is one
of the key requirements for design and
engineering systems For example, an OEM
decides to introduce a variant of an existing
platform of one of its subsidiaries in Europe
to another
incorporating some customizations. In such
a scenario the Asian subsidiary of the OEM
will need access to design systems of the
European subsidiary. In addition the design
systems of the Asian subsidiary will also need
to share information with production planning,
purchase, finance as well as preferred suppliers
design systems.
The various entities involved in this
scenario are shown Figure 2 which depicts a
typical integration challenge in supply chain
subsidiary in Asia after collaboration and the way it can be addressed
using an SOA based integration platform. In most
scenarios, the design systems of a manufacturer
in different geographies will be different. In
the stated scenario data needs to be moved from
design system in Europe to a design system
in Asia.

119
 Design system in Europe provides
application interfaces for data extraction as
XML. However, data needs to be provisioned as
a flat file over FTP to the design system in Asia.
In order to transfer data between concerned
applications, individual pieces of code can
be written that invoke API in Europe design
system that obtains data as XML, formats it
to a flat file and saves it in a required location
in Asia subsidiary’s system. Such point to
point connection based approach will require
application to invoke the native API (e.g.,
JAVA). However, any other application that
desires to get the same data will need to replicate
the same function.
The solution lies in exposing native
API as services (e.g., web services using
SOAP over HTTP) that adhere to enterprise
level data formats. Once created, these services
can be used by any other applications. In the
absence of required API, data services can be
directly generated from the concerned data
sources.
Data services are generated by
definition of standards-based query in a
centralized data services platform and exposed
either using standards based interfaces like
web services or some API. The retrieved data
is translated to the required flat file format __
where the source and destination ontology
and required translations are drawn from a
metadata repository __ and transmitted to
a physical location using enterprise service
bus (service at integration layer). This approach
allows standards-based definition of service
interface that exchanges data in mutually
understood format and introduces a mediation
layer that handles translation, routing and
transport bridging (HTTP to FTP, in this case).
In certain scenarios there might
be a need for designers at Asian subsidiary to
120
lookup for a particular part in the European
OEM’s design systems. However, due to
different naming conventions it might be
difficult to locate the exact required part. In
such scenarios, shared data services platform
can use semantic definitions in metadata
repository to perform lookup for parts
that follow different naming conventions.
The design system at Asian OEM
needs to share data with other systems like
manufacturing systems, prototyping systems
and purchasing systems, all of which require
different data formats over multiple protocols.
Also there is a need to share design data
with preferred supplier design system using
STEP (ISO 10303) for PLM (Standard for the
Exchange of Product Model Data) [10].
A service bus can be used for
achieving the desired integration and for
provisioning of federated security while
exchanging data with preferred suppliers’
systems located across different domains.
The planning system at Asian OEM requires
data from one or more manufacturing
systems. However, these manufacturing
systems do not provide any API for retrieval
of such data. Traditional means of achieving
the result would be by writing custom JDBC or
application API centric code (for e.g., ABAP in
case of SAP). However, the resulting point to
point connections are hard to maintain and are
replicated for every new application.
The solution lies in definition of
distributed queries in standards-based manner
in SDS platform and exposing these queries as
services to be re-used. SDS platform takes care of
distributed connection management, transaction
management, caching and tool based generation
of service interfaces.
Finally the submission of purchase
order from Asian manufacturers’ purchase
system to suppliers’ purchase system can be 5. I2 Operational data store. I2 technologies
seen as process level service that internally LTD. http://www.i2.com/assets/pdf
drives manufacturing process at suppliers end /PDS_ods_v61_pds7262_091304.pdf.
and provides asynchronous callback handlers 6. John Leslie King and Kalle Lyytinen,
to service consumers for checking status of Automotive Informatics: Information
purchase order. Technology and Enterprise
Transformation in the Automobile
CONCLUSION Industry. http://www.si.umich.edu/
Collaboration in auto supply chain involves ~jlking/Auto-MIT-Final.pdf.
participation of multiple stakeholders across 7. V Niranjan et.al., Shared Data
the supply chain towards sharing information Services: An Architectural Approach.
at different levels of supply chain management http://portal.acm.org/citation.
activities. Traditional means of point to point cfm?id=1090954.1092078.
application integration falls short of achieving 8. Bijoy Majumdar et.al., SOA
the objective of collaboration and is not suited Security Modelv2.0. http:/whatis.
for agile business processes. SOA provides a techtarget.com/tip/0,289483,sid92_
new approach to address integration needs gci1168738,00.html.
in auto supply chain management towards 9. Enterprise service bus, http://www-
enabling collaboration. 3 0 6 . i b m . c o m /s o f t w a r e /i n f o 1/
An SOA based integration platform websphere/index.jsp?tab=landings/
allows standards based integration of multiple esbbenefits.
types of interfaces that communicate using 10. STEP (ISO 10303). http://www.
different data formats over different protocols. steptools.com/library/standard/.
Hence such a platform is ideally suited to 11. The American Automotive Industry
handle fluid integration requirements in a Supply Chain- In the Throes of a
collaborative supply chain. Rattling Revolution. http://www.ita.
doc.gov/td/auto/domestic/
REFERENCES SupplyChain.pdf.
1. Laurie Sullivan, Auto-Parts Makers 12. Luis Araujo, Modularity, Systems
Open Their Networks, Information Integration and Supply Chain
Week, Available at http://www. Leadership. http://www.ht2.org/
agile.com/news/2004/infoweek_ conference/pdf/SYM1.pdf.
041904.pdf. 13. Ricardo UL Ltd, Skills4Auto Ltd. Vision
2. Ford T Model. http://en.wikipedia. of UK Automotive industry in 2020.
org/wiki/Ford_Model_T. www.ricardo.com/media/
3. Peter Wad, Workers in the Supply Chain medialibrary/Vision_for_the_UK_
of the Global Automobile Industry. Automotive_Industry_in_2020.pdf.
http://www.amrc.org.hk/5404.htm. 14. Where is the auto industry headed?
4. Kanban, University of Cambridge. http://www.deloitte.com/dtt/article/
http://www.ifm.eng.cam.ac.uk/ 0,1002,cid%253D120316,00.html.
dstools/process/kanban.html. 15. Andrew Cummins, Supply Chain vs.

121
Supply Chain - automotive industry www.findarticles.com/p/articles/
distribution - Brief Article. http:// mi_m3012/is_6_181/ai_76855607.

122
 THE LAST WORD

“Shrinking the Expanding

Supply Chain”

Competitiveness of global enterprises will depend on

the exchange and consumption of supply chain
information. This, the consulting editor, Dr. Badri
Devalla feels would help mask the distances in inventory
movement and manage the risk in financial flows

G lobal Supply Chains (GSCs) are one of

the shining examples of how business is
done in a flat world. GSCs have allowed for
not only manufacturing to be based in Asia,
but also sourcing and procurement from there.
Global operations have also contributed to
improvements in the distribution of consumer
and industrial goods in consumption regions
of the U.S. and Europe. Automotive supply
chains have shown how one can be lean and
profitable in high wage markets like the U.S.
Supply chains today are much longer
and have expanded to include more partners,
than even just a decade ago. We have seen
that many of the new trends and strategies
foster a federated supply chain management
to effectively mask the distances between the
suppliers and consumers. Supply chains can
be shrunk through an efficient movement of
resources along the three classic dimensions
of the supply chain: faster movement of larger
amounts of inventory across larger distances,
information being exchanged easier through
internet-enabled structures, money moved
easily across borders and global partners.
We hope you enjoyed the topics and
the papers we have selected for this issue. We
have presented a select set of contributions
around new trends, strategic responses and
perspectives on shrinking the expanding
supply chain along these dimensions:
Inventory: How can you continuously
lower fixed costs in carrying inventory from
procurement to manufacturing to distribution?
How do you improve inventory performance
in a world driven by energy efficiency and
environmental regulations?
Information: How do you leverage information
to work effectively in the partner ecosystem?
How can you monetize supply chain

123
information on demand, supply, inventory and
logistics?
Financial flows: How can you improve working
capital performance through innovative
techniques to manage financial flows? How do
you manage risk and liability across a global
supply chain and complex regulations?
Global enterprises will continue to
seek innovative techniques and strategies that
look outward toward their partners to evolve
their supply chains to continue to lower
inventories, get better visibility of demand
and supply, and improve working capital
performance. GSC management still needs
significant upgrades to be more responsive
to demand variability in an ecosystem that is
increasingly about pulling components and
vendor managed inventories.
New techniques in inventory
management and information management
are needed to make the supply chain less
opaque. Inventory optimization techniques
that consume supply chain information for
better inventory visibility and control will
124
become more critical. Competitiveness in a
supply chain spread around the globe will
also engender new strategies to maximize
the performance of financial flows - an area
that Stanford Professor Warren H. Hausman
notes has not seen much change since
the 1970s.
Please share your thoughts on the
papers presented in this issue and also on any
trends and strategies we may have missed.
About the Author
Dr. Badri Devalla is a Senior Principal
Consultant with Infosys’ High Tech and Discrete
Manufacturing business unit. His current focus
is in designing and delivering Supply Chain
solutions. He is particularly interested in business
value realization around the creation, exchange and
consumption of supply chain information. As an
RFID expert, he has presented his ideas in several
forums, including Knowledge@Wharton. He has
several years of experience as a Systems Architect,
a Product Manager and a Consultant in the High
Tech industry. He has a PhD in Computer Science
from Texas A&M University and can be reached at
B_Devalla@Infosys.com.
 Index
.NET 115-116 Financial Supply Chain Management, also
Accounts 27, 31, 34, 84, 88, 90 FSCM 28, 31-32, 83, 87, 91
Payables, also AP 27, 31, 34, 84, 88, 90 Flat world 3, 5, 8, 104, 123
Receivables, also AR 27, 31, 34, 90 Ford 10-11, 113, 121
Advanced Planning and Scheduling, also Full truckload, also FTL 67-68
APS 18, 25 General Electric, also GE 4, 31
Advance Ship Notice, also ASN 65 Global outsourcing 28, 33, 102
Amazon 5, 13, 23 Hewlett Packard, also HP 10-11, 23, 75
Application Programming Interfaces, also HTTP 115, 119-120
API 107, 116-118, 120 Hurricane Katrina 94
Automated Clearing House, also ACH 86 Innocentive 4, 13
BMW 10 Intel 5
Bill of Materials, also BOM 54, 58 Inventory 15-25, 58, 114, 124
Cisco 4, 6, 10 Control 16, 18, 58, 114
Collaborative Planning, Forecasting and Optimization 15-25, 124
Replenishment, also CPFR 105, 107-112 JMS 115, 119
Crude 34-36, 38-42 Just-in-time, also JIT 6-7, 23, 64, 95, 114
Exploration 35-36, 38, 42 Less-than-truckload 64-68
Extraction 34, 39, 41 Low cost country sourcing 21, 28-29, 33, 78, 93
Loading 36, 38 Maintenance, Repair and Operations, also
Production 36, 42 MRO 30, 78, 80, 97
Selection 36, 40 Marketics 38, 40, 42
Unloading 38 Master Data Management, also MDM 12, 44, 75
Demand Driven Supply Network, also DDSN 5 Material Requirements Planning, also MRP 15,
Dell 5, 10-11, 16, 23, 32, 34 18, 22, 25, 45, 57-58, 60
Dow 94 Moore’s Law 6
EJB 115 On Time In Full, also OTIF 55
Electronic Data Interchange, also EDI 65, 79, 89, Original Equipment Manufacturer, also OEM 6,
106, 112, 115 8, 10, 55, 58, 60, 100, 114-115, 118-120
Electronic Invoice Presentment and Payment, Philips 94
also EIPP 28, 31, 34, 86-87 Procter and Gamble, also P&G 4,13, 16
Enterprise Cost Management, also ECM 71-73, RMI 115
76, 80-81, Rossettanet 79
Enterprise Information Integration, also EII 38, Risk
107, 112 Business 7
Enterprise Service Bus, also ESB 117, 120-121 Pooling 23-24
Ericsson 9, 94 Supply chain 7-8, 12-13, 94

125
Service Oriented Architecture, also SOA 12, 79, Demand Planning 68
113, 116-119, 121 EAI 115
Single Node Multi-Processors, also SMP 37 Event-Management 12
Six Sigma 31 Implementation 98
STAR 115 IT 51
Supplier 99, 101 Machine 24
Quality Audits 99 OLAP 75
Scorecard 99 Operational Efficiency 100
Self-assessment 101 Power 25
Supplier Managed Inventory, also SMI 17, 23-24 Process 86
Supply Chain Operations Reference Model, also Simulation 42
SCOR Model 101 Software 99
Supply Chain Risk Management, also Strategic 100
SCRM 5, 103 Supply Chain 56
T Model 113, 121 Vendor Managed Inventory, also VMI 32,
Transportation Management System, also TMS 56, 107
64, 68-69 Voluntary Inter-industry Commerce Standards,
Tools 12, 24-25, 42, 51, 56, 68, 75, 86, 89, 98-100, also VICS 105, 107-108, 111-112
115 Warren H. Hausman 34, 126
Analytic 100 XML 79, 88-89, 105-112, 115, 119-120
Connectivity 89 XQuery 105, 110, 112, 116

126

BUSINESS INNOVATION through TECHNOLOGY
Editor
George Eby Mathew
Consulting Editor
Badri Devalla PhD
Associate Editor
Srinivas Padmanabhuni PhD
Copy Editor
Anupama Gummaraju
Graphics/Web Editors
Ramesh Ramachandran
Sanjay Eknath Nayak
ITLS Program
Ajay Kolhatkar
Marketing Manager
Vijayaraghavan T S
Production Managers
Sudarshan Kumar V S
V H Suresh Kumar
How to Reach Us:
Email:
SETLabsBriefings@infosys.com
Phone:
+91-20-39167531
Fax:
+91-20-22932832
Post:
SETLabs Briefings,
B-19, Infosys Technologies Ltd.
Electronics City, Hosur Road,
Bangalore 560100, India
Subscription:
vijaytsr@infosys.com
Rights and Permission:
george_mathew@infosys.com
Reprints:
george_mathew@infosys.com
SETLabs Briefings
Editorial Office: SETLabs Briefings, B-19, Infosys Technologies Ltd.
Electronics City, Hosur Road, Bangalore 560100, India
Email: SetlabsBriefings@infosys.com hhttp://www.infosys.com/technology/SETLabs-briefings.asp
SETlabs Briefings is a quarterly published by Infosys’ Software Engineering
& Technology Labs (SETLabs) with the objective of offering fresh
perspectives on boardroom business technology. The publication aims at
becoming the most sought after source for thought leading, strategic and
experiential insights on business technology management.
SETLabs is an important part of Infosys’ commitment to leadership
in innovation using technology. SETLabs anticipates and assesses the
evolution of technology and its impact on businesses and enables Infosys
to constantly synthesize what it learns and catalyze technology enabled
business transformation and thus assume leadership in providing best of
breed solutions to clients across the globe. This is achieved through research
supported by state-of-the-art labs and collaboration with industry leaders.
Infosys Technologies Ltd (NASDAQ: INFY) defines, designs and delivers
IT-enabled business solutions that help Global 2000 companies win in a
flat world. These solutions focus on providing strategic differentiation
and operational superiority to clients. Infosys creates these solutions
for its clients by leveraging its domain and business expertise along
with a complete range of services. With Infosys, clients are assured of a
transparent business partner, world-class processes, speed of execution
and the power to stretch their IT budget by leveraging the Global Delivery
Model that Infosys pioneered. To find out how Infosys can help businesses
achieve competitive advantage, visit www.infosys.com or send an email to
infosys@infosys.com
© 2007, Infosys Technologies Limited
Infosys acknowledges the proprietary rights of the trademarks and product names of the other companies
mentioned in this issue. The information provided in this document is intended for the sole use of the recipient
and for educational purposes only. Infosys makes no express or implied warranties relating to the information
contained herein or to any derived results obtained by the recipient from the use of the information in this
document. Infosys further does not guarantee the sequence, timeliness, accuracy or completeness of the
information and will not be liable in any way to the recipient for any delays, inaccuracies, errors in, or omissions
of, any of the information or in the transmission thereof, or for any damages arising there from. Opinions and
forecasts constitute our judgment at the time of release and are subject to change without notice. This document
does not contain information provided to us in confidence by our clients.
Authors featured in this issue
ANUPAM KULSHRESHTHA
Anupam Kulshreshtha PhD, is a Consultant with the Analytics Services practice of Infosys’ Domain Competency Group. He
can be reached at Anupam_K@infosys.com.
ARNAB BANERJEE
Arnab Banerjee is a Consultant with Infosys Technologies Limited, Enterprise Solutions. He has considerable experience
in large scale IT projects involving ERP and Planning Solutions in Oracle Applications. He can be contacted at Arnab_
banerjee01@infosys.com.
ASHISH KUMAR TEWARY
Ashish Kumar Tewary PhD, is Principal Consultant with Infosys’ SAP Practice. He manages Infosys’ portfolio of solutions on
SAP technology platform for RFID and Service-Oriented Architecture. He can be reached at ashish_tewary@infosys.com.
ASHWINI KUMAR JEKSANI
Ashwini Kumar Jeksani is a Software Engineer with the Web Services/SOA Center of Excellence in SETLabs, Infosys
Technologies Limited. He can be reached at Ashwini_Jeksani@infosys.com.
BHARANI SHIVARAJA GOVINDASAMY
Bharani Shivaraja Govindasamy is an Associate Consultant with the supply chain management practice at Infosys. His interests
are in demand planning where most of his research work has been spent. He can be contacted at bharanishivaraja_g@infosys.com.
G. V. GANESH
G. V. Ganesh is a Principal Consultant with Infosys Supply Chain practice. His areas of expertise include Procurement,
Strategic Sourcing and Collaborative Supply Management. He can be reached at Gavarapet_Ganesh@infosys.com.
GREG SCHEUFFELE
Greg Scheuffele is a Principal and Solutions Manager in Infosys’ High Tech and Discrete Manufacturing practice. He
manages Infosys’ portfolio of solutions for inventory design, optimization, and replenishment. He can be contacted at Greg_
Scheuffele@infosys.com.
HARIHARAN SUBRAMANIAN NOORANI
Hariharan Subramanian Noorani is a Consultant with the Domain Competency Group in the retail vertical. He can be
reached at Hariharan_noorani@infosys.com.
KRISHNENDU KUNTI
Krishnendu Kunti is a Senior Technical Specialist with the Web Services CoE, SETLabs, Infosys. He can be reached at
krishnendu_kunti@infosys.com.
NARAYANAN SAMPATH
Narayanan Sampath is an Associate Vice President with Infosys BPO and is responsible for managing outsourcing operations.
He has rich experience in different industries across various functions — design, manufacturing, supply chain mangement
and implementing SAP solutions. He can be contacted at narayanan_sampath@infosys.com.
RAJ JAYARAMAN
Raj Jayaraman is a Principal with Infosys High Tech & Discrete Manufacturing Business Unit. He has led several
transformational supply chain engagements. He can be reached at rajesh_jayaraman@infosys.com.
RAJIB SAHA
Rajib Saha is a Senior Consultant with manufacturing and supply chain practice in the Domain Competency Group. He can
be contacted at rajib_saha@infosys.com.
SABITHA VUPPALA
Sabitha Vuppala is a Principal with Infosys’ Banking & Capital Markets practice. She can be contacted at Sabitha_
vuppala@infosys.com.
SANDEEP KUMAR
Sandeep Kumar leads the Manufacturing and Supply Chain group at the Domain Competency Group in Infosys. He can be
reached at Sandeep_kumar@infosys.com.
SHUBHASHIS SENGUPTA
Shubhashis Sengupta PhD, is a Principal researcher at Infosys working in High performance, Grid and service oriented
computing. He has several years of rich research and consultancy experience. He can be reached at shubhashis_
sengupta@infosys.com.
SIVA PADMANABHAN
Siva Padmanabhan leads supply chain related service offerings for Infosys Consulting. His areas of expertise are in
management of large business and IT programs, strategic Sourcing and Procurement transformation. He can be contacted
at siva_padmanabhan@infosys.com.
SUJATA BANERJEE
Sujata Banerjee is a Principal with Infosys’ Banking & Capital Markets practice. She can be contacted at Sujata_banerjee@infosys.com.
SUMESH GEORGE
Sumesh George is a Principal Consultant with Retail Industry Solutions Consulting Group. His areas of expertise include
Business Strategy, BPR, Process Improvement, Supply Chain Performance Management. He can be reached at Sumesh_
george@infosys.com.
SUMIT KUMAR BOSE
Sumit Kumar Bose PhD, is a Junior Research Associate at Infosys working in the area of High performance computing. He
can be reached at sumit_bose@infosys.com.
SURESH PRAHLAD BHARADWAJ
Suresh Prahlad Bharadwaj is a Principal Consultant in the Retail Industry Solutions Consulting Group. He can be reached
at Suresh_bharadwaj@infosys.com.
TERANCE DIAS
Terance Dias is a technical specialist with the Web Services/SOA Center of Excellence in SETLabs, Infosys Technologies
Limited. He can be reached at terance_dias@infosys.com.
V. S. SRIVIDHYA
V. S. Srividhya is a Project Manager with the High Risk Projects Group at the Corporate Practice of Infosys. She can be
contacted at srividhya_vs@infosys.com.
 Subu Goparaju “At SETLabs, we constantly look for opportunities to leverage
Vice President
technology while creating and implementing innovative business
and Head of SETLabs
solutions for our clients. As part of this quest, we develop engineering

methodologies that help Infosys implement these solutions right

first time and every time”.

For information on obtaining additional copies, reprinting or translating articles, and all other correspondence,
please contact:
Telephone : 91-20-39167531
Email: SetlabsBriefings@infosys.com

© SETLabs 2007, Infosys Technologies Limited.

Infosys acknowledges the proprietary rights of the trademarks and product names of the other
companies mentioned in this issue of SETLabs Briefings. The information provided in this document
is intended for the sole use of the recipient and for educational purposes only. Infosys makes no
express or implied warranties relating to the information contained in this document or to any
derived results obtained by the recipient from the use of the information in the document. Infosys
further does not guarantee the sequence, timeliness, accuracy or completeness of the information and
will not be liable in any way to the recipient for any delays, inaccuracies, errors in, or omissions of,
any of the information or in the transmission thereof, or for any damages arising there from. Opinions
and forecasts constitute our judgment at the time of release and are subject to change without notice.
This document does not contain information provided to us in confidence by our clients.