Innovative Supply Chain

Strategies

Rajesh Piplani, Ph. D.

Director, Center for Supply Chain Management
Nanyang Technological University
Agenda
• Background
• Innovative SC Strategies
• Supply Chain Selection
– Sony digital camera production
• Supply Chain Realignment
– Li & Fung

Rajesh Piplani, Ph. D. © 2007 2

 Supply Chain Drivers
• Globalization
• Specialization
– Out-sourcing
• Latest example: McDonald’s drive-thru order processing
• Product variety
• Closed loop supply chain management

Rajesh Piplani, Ph. D. © 2007 3

Sites: Assembly--Assembly Plants
Plant--Manufacturing Plants
DC--Distribution Centers
Markets--Wholesale Dealers
Ports--Shipping Ports
Rajesh Piplani, Ph. D. © 2007 4
Vendors--Raw Material Suppliers
Product variety
Nokia launched a total of 32
different hand-phone models
in 2006 in all markets.

In Aug 2006 in Beijing, HP

launched 70 different printing
and imaging products.

Rajesh Piplani, Ph. D. © 2007 5

 Closed Loop SCM
• Selling a service rather than a product
• Disposable camera
• Photo copier
• Hand-phone
• Returns management
• Warranty repair
• 3Rs (Refurbish, Recover, Recycle)

Rajesh Piplani, Ph. D. © 2007 6

Agenda
• Background
• Supply Chain Strategies
• Supply Chain Selection
– Sony digital camera production
• Supply Chain Realignment
– Li & Fung

Rajesh Piplani, Ph. D. © 2007 7

 Innovative SC Strategies
• Postponement
• Cross Docking
• Demand-driven Supply Chains

Rajesh Piplani, Ph. D. © 2007 8

Without any Commonality and
Postponement
Operation Buffer

Model A

Model B

Model C

Rajesh Piplani, Ph. D. © 2007 9

 With Postponement and
Commonality
Operation Buffer

Incremental investment
for producing common components
and exploiting economies of scale
Model A

Model B

Model C

Potential savings due to

reduction of inventory

Rajesh Piplani, Ph. D. © 2007 10

 With Postponement and Commonality
Operation Buffer

Incremental investment Manufacturing

for producing common components plant
and exploiting economies of scale
Model A

Model B

Model C

Potential savings due to Distributor

reduction of inventory or
Retailer

Driven Driven
by aggregate Make to Stock Make to Order by individual
forecast demand
Rajesh Piplani, Ph. D. © 2007 11
Push-Pull boundary
 Postponement: Case Study

• Benetton
– Re-configuration of production sequence

Rajesh Piplani, Ph. D. © 2007 12

 Case Study
Problem
• Benetton’s knitwear includes more than 500 color and style
combinations
• Store owners required to place orders for wool sweaters up to seven
months in advance
• In fashion industry consumer preferences change rapidly, but because
of long lead times company had little flexibility to respond to changing
tastes of consumers
Solution
• Traditionally, wool is dyed before it is knitted, Benetton developed a
process where garments could be dyed after knitting (increased
production cost by 10%)
• Uncolored sweaters are made to forecast, but dyeing takes place as a
reaction to customer demand

Rajesh Piplani, Ph. D. © 2007 13

DATA RED BLUE GREEN YELLOW
Retail price $50 $50 $50 $50
Production cost (option 1) $20 $20 $20 $20
Production cost (option 2) $22 $22 $22 $22
Salvage value $10 $10 $10 $10 SUMMARY OPTION 1 OPTION 2
Mean demand 1,000 1,000 1,000 1,000 Quantity 5349 4524
Std dev of demand 500 500 500 500 Profit $94,578 $98,092
Overstock 1647 715
OPTION 1 RED BLUE GREEN YELLOW Understock 298 190
Optimal service level 75% 75% 75% 75%
Optimal purchase quantity 1,337 1,337 1,337 1,337

Expected profit $23,644 $23,644 $23,644 $23,644

Expected overstock 412 412 412 412
Expected understock 75 75 75 75

OPTION 2 AGGREGATE
Optimal service level 70%
Mean demand 4,000
Std dev of demand 1,000
Optimal purchase quantity 4,524
Expected profits $98,092
Expected overstock 715
Expected understock 190

Rajesh Piplani, Ph. D. © 2007 14

 (Tailored Postponement)
– Use lower-cost production to satisfy the predictable part of a
product’s demand and use postponement to meet the portion of
the demand that is uncertain
– Benetton requires retailers to commit 70% of their volume about
seven months in advance and the remaining orders can be
placed much closer to or even after the start of the season
– Firm orders are subcontracted to low-cost sources that have long
lead times and manufactured using option 1 (knitted using
colored thread)
– Uncertain part of the demand is produced by Benetton in its
own flexible facility using option 2 where safety stock is held as
Greige goods and manufactured on demand (10% more
expensive but short lead times)
Rajesh Piplani, Ph. D. © 2007 15
 Forecasting
Each data point represents the forecast and the actual
Actual total sales season sales for a particular item (at the style-color level).

4000 4000 4000

3500 3500 3500

3000 3000 3000

2500 2500 2500

2000 2000 2000

1500 1500 1500

1000 1000 1000

500 500 500

0 0 0
0 500 1000 1500 2000 2500 3000 3500 4000 0 500 1000 1500 2000 2500 3000 3500 4000 0 500 1000 1500 2000 2500 3000 3500 4000

Updated Forecast Updated Forecast

Initial Forecast after observing 20% of sales after observing 80% of sales

Increase in forecast accuracy decreases both the overstock and

understock quantity andRajesh
improves thePh.
Piplani, firm’s
D. ©profits
2007 16
 Insights on Postponement Strategy
• Large variety of products (customization) and long lead times
make forecasting and inventory management a difficult task
• Demand for the products is not positively correlated and is of
about the same size for all the products
• May reduce the overall profits for a firm if a single product
accounts for majority of its demand
• Tailored postponement will typically improve profits only if
– the uncertain part of the demand is postponed and
– the predictable part of the demand is produced at a lower cost, without
postponement

Rajesh Piplani, Ph. D. © 2007 17

 Cross Docking is a Simple concept
• Cross docking:
– speeds delivery to customers,
– reduces the cost to process an order, and
– reduces inventory, all at the same time.
• The concept is simple:
– When the bar code label on an incoming carton is
scanned, the WMS automatically allocates that product
to a specific store.
– The carton is then automatically conveyed and sorted to
a shipping lane with other cartons going to that store.

Rajesh Piplani, Ph. D. © 2007 18

 Cross Docking is cost effective
• It is also cost-effective: One retailer reduced
the cost per carton handled from $1.40 to $0.14
by implementing an automated cross docking
system, according to Dematic.

Rajesh Piplani, Ph. D. © 2007 19

Cross Docking requires coordination
• While the concept is simple, the process itself relies on the
coordination of a variety of real-time systems, including:
• Advance ship notices (ASNs)
– from suppliers so the system has visibility into what's coming in,
• Information from a transportation management system (TMS)
– to identify when it will arrive and when the outbound truck is
scheduled to leave,
• Demand information from inventory replenishment systems
– to know what stores are in need of the inventory,
• A warehouse control system
– to manage the sorting of the product, and
• A WMS
– to orchestrate the procedure from the receiving dock to the right
shipping lane.

Rajesh Piplani, Ph. D. © 2007 20

 Cross Docking: inventory transfer
• As customers become more accustomed to operating in real time, they
are using that information to enable new processes
– the combination of reverse logistics and cross docking.
• Using real-time point-of-sale and store-replenishment systems,
corporate planners find that they've sold out of 34-inch sized jeans in
one store, while they have a surplus in another store.
– In the old days, planner would get on the phone and manually
arrange for one store to ship their surplus to another store. Now,
that process can be automated.
• A retailer's replenishment system can order
– a real-time transfer of excess inventory from a store to a DC.
• When that inventory arrives at the DC
– it can be cross docked and put on the next truck to the store that
needs that inventory, just like a delivery from a supplier.
Rajesh Piplani, Ph. D. © 2007 21
 Cross docking: allocation postponement
• With real-time visibility into the contents of
containers coming into port along with inventory
levels in DCs and stores
– retailers can now postpone the pre-allocation of
inventory on a cargo ship until the very last moment.
• Just a few years ago, they had to pre-allocate that
inventory before it was loaded into a container and
put on a cargo ship.
• Now, they can wait to allocate until it hits the
dock, then it can be cross docked to a store based
on demand. Rajesh Piplani, Ph. D. © 2007 22
 Cross Docking: Examples

• Wal-Mart DCs

• PSA terminals

Rajesh Piplani, Ph. D. © 2007 23

 Demand Driven Supply Chains
• Demand Driven Supply Chains
– Don’t build to forecast, but build-to-order (BTO)
• Business Partner Integration
– Integration (with partners) by connecting the IT
systems
– This facilitates real-time visibility
– Which in turn leads to
• Synchronization of activities
• Reduction in cycle times
• Elimination of large buffers of inventory
Rajesh Piplani, Ph. D. © 2007 24
 Supply Chain planning
and collaboration Integration within
Rationalized parts enterprise (hub)
and products

Demand-driven Integration of
Supply-chain Business partners

Visibility for extended

Fast response enterprise
manufacturing

Connection to extended
Fast response enterprise (portal)
logistics

Rajesh Piplani, Ph. D. © 2007 25

 BTO Supply Chain
• BTO Supply chains lead to reduction in
inventory.
• But then , the Supply Chain has to be agile
enough to react to the variability inherent in
customer demand.
• This is where integration helps.

Rajesh Piplani, Ph. D. © 2007 26

 New Supply Chain Vision
Old Structure: Build to forecast

Forecast driven Demand driven

Push Push Customers
Batch processing Pull
Channel
Raw material Finished-goods Partners
inventory

New Vision: Build to Order

Forecast Demand driven

driven Pull Pull Customers
Flexible
Manufacturing Pull
Channel
Vendor’s Raw Finished-goods Partners
material Inventory inventory
Rajesh Piplani, Ph. D. © 2007 27
Agenda
• Background
• Supply Chain Strategies
• Supply Chain Selection
– Sony digital camera production
• Supply Chain Realignment
– Li & Fung

Rajesh Piplani, Ph. D. © 2007 28

Right supply chain for your product!

• Sony digital camera and camcorder

production

Rajesh Piplani, Ph. D. © 2007 29

 Sony Digital
• In 1995, facing pricing pressures, Sony moved its
Digital Camera and Camcorder manufacturing
operations to China.
• Main competitors: Canon and Olympus.
• Expensive products. Competition based mainly on
price.
• Objective (of the move)
– To lower product’s assembly costs

Rajesh Piplani, Ph. D. © 2007 30

 Japan

China

USA

Assembly Center
Suppliers
Components Flow
Final Products Flow
Rajesh Piplani, Ph. D. © 2007 31
 Digital Camera market
• Over the next few years, as prices dropped (due to advances in
technology and product becoming mass market), competition
moved to product variety and advanced technology.
• Companies gained market share by introducing newer and more
technologically advanced products faster and faster.
• Life Cycle for these products dropped to two-three months.
• These products contain high-tech parts, such as full-color LCD
screen, precision optical machinery parts, digital memory and
CCD.
• Japanese digital cameras hold almost 90% of the global market
share (Major markets: Japan and USA).
– Japanese manufacturers possess the advanced technology to
manufacture its components.

Rajesh Piplani, Ph. D. © 2007 32

 Made in Japan again!
• In July 2002, Sony shifted its digital camera
and camcorder manufacturing back to
Japan.
• According to Yoshihiro Taya (VP, Sony),
move to Japan will cut the lead time for
camcorder production in half.

Rajesh Piplani, Ph. D. © 2007 33

 Japan

China

USA

Assembly Center
Suppliers
Components Flow
Final Products Flow
Rajesh Piplani, Ph. D. © 2007 34
What pulled Sony out of China?
• Sony’s R&D center, critical suppliers and
customer service are all located in Japan.
• Sony’s R&D center needs to be able to quickly
gain feedback from customers and incorporate
that into the next version.
– Think product development for short life-cycle
innovative products.
• Most of Sony’s Japanese suppliers were
hesitant to move their manufacturing facilities
to China.
– The Chinese govt. requires that all foreign
companies train their local partners in the advanced
technology associated with their products.
Rajesh Piplani, Ph. D. © 2007 35
 Opportunities for CT reduction
Can be improved in Can be Improved in
Japan China
Materials planning and
Yes No
scheduling
Purchase order cycle Yes No
Inbound transportation Yes No
Material receipt/inspection Yes No
Manufacturing Processes Yes Yes
Customer order processing Yes No
Warehousing operations Yes Yes
Outbound transportation Yes No
Reverse Logistics Yes No
Rajesh Piplani, Ph. D. © 2007 36
 Lessons learned?
• Certain supply chain factors may be even more
important than manufacturing costs:
– Product cycle time
– Proximity to suppliers
– Speed to market
• Different products require different supply chain
strategies.
– For innovative products (such as digital cameras and
camcorders), cycle time reductions contribute more to the
company’s competitiveness than physical cost reductions.
– Advantage of being able to respond quickly to emerging
customer demand may offset the disadvantage of high labor
costs. Rajesh Piplani, Ph. D. © 2007 37
Rajesh Piplani, Ph. D. © 2007 38
Agenda
• Background
• Supply Chain Strategies
• Supply Chain Selection
– Sony digital camera production
• Supply Chain Realignment
– Li & Fung

Rajesh Piplani, Ph. D. © 2007 39

 Supply Chain Realignment
• Companies will have to realign their supply chains
– interconnect their SC systems with their partners’
systems
– adopt processes that help off-load (if required)
production responsibilities to contract manufacturers
– allow supply chain partners to collaborate
• Better communication and flexibility would be the
key to success for global companies

Rajesh Piplani, Ph. D. © 2007 40

 Supply Chain Realignment
• Well-connected assets (facilities) aid visibility and allow
participation in highly connected business networks of
tomorrow.
• Companies need visibility into their contract service
providers’ operations.
– Flextronics allows its customers (OEMs) to track their order status
on the shop floor.
– SMTC’s planning systems let customers like Dell run what-if
scenarios over the web. Dell can use this tool to determine if
SMTC’s Shanghai plant can produce enough motherboards to meet
the surge in PC demand in China. Dell can even evaluate the
performance of SMTC’s suppliers.

Rajesh Piplani, Ph. D. © 2007 41

 Supply Chain Realignment
• Competition would force companies to focus on their core
competency (out-source other activities), and be able to
plug into multiple networks.
• Specialization would also help focus on customer demand
characteristics.

Rajesh Piplani, Ph. D. © 2007 42

 Action items for the future
• Web-centric demand planning
– allow constituents (both internal and external) to collaborate on the
same data to generate viable forecasts
• Improving product development process
– using web-based tools to do collaborative design
• Production for late-phase differentiation
– To support make-to-order production, companies should engineer
their plants for late-phase differentiation somewhere in the supply
chain
• Tracking performance of supply chain partners
– Using extended supply chain visibility to monitor and assess the
partners’ performance in relation to the customer demand

Rajesh Piplani, Ph. D. © 2007 43

 Virtual organizations
• To meet the demand swings, companies can turn to: E-
marketplaces, capacity clearinghouses or virtual
organizations.
• Virtual manufacturers (called Dot-builds by Forrester
Research) or 4th party manufacturers (4PMs).
• E-business network (domain specific) of component
suppliers (in Asia), assembly capacity (in Brazil), and
distribution companies (in Europe).
• (Get) build anything to your design specs.

Rajesh Piplani, Ph. D. © 2007 44

 Reaching into the Supply Chain
• Li & Fung (Hong Kong): a virtual manufacturer for the
garment industry.
• Example:
– Past cycle times three months.

Rajesh Piplani, Ph. D. © 2007 45

 Cycle Time = 3 Months
1 WK Yarn 1 WK Weaving 1 WK
Order Arrival
Supplier

2 WK
Garment
Ship Out 1WK 1 WK 2 WK Dyeing 1WK
Manuf. 2 WK

Rajesh Piplani, Ph. D. © 2007 46

 Reaching into the Supply Chain
• Example:
– To shrink the delivery time, Li & Fung goes upstream to
organize production.
– “Limited” to order 10,000 garments, don’t know the color or
style. Lead time to deliver 5 weeks.

Rajesh Piplani, Ph. D. © 2007 47

 Reaching into the Supply Chain
• L&F’s new approach (using a shell order)
– L&F reserves undyed yarn with the yarn supplier.
– And locks up capacity at the mills for weaving and dyeing;
arranges for delivery of the yarn to them, and they have 4
weeks to manufacture.
• It is called reaching into the SC to shorten cycle times.
• What makes their approach work?

Rajesh Piplani, Ph. D. © 2007 48

 Cycle Time = 5 Weeks
Yarn Weaving 1 WK
Order Arrival
Supplier

Garment
Ship Out 1WK Dyeing 1WK
Manuf. 2 WK

Rajesh Piplani, Ph. D. © 2007 49

 In conclusion…
• Companies will focus more and more on
their core competencies and outsource all
other operations.
• Critical to choose the right supply chain for
your product.
• Technology will be used more and more to
streamline, optimize and make SC
operations more efficient.

Rajesh Piplani, Ph. D. © 2007 50

 Q&A

Rajesh Piplani, Ph. D. © 2007 51