Bounded Rationality

Download Report

Transcript Bounded Rationality

Welcome
Strategies of Network Companies
Jonathan D. Wareham
[email protected]
Agenda



When firms cooperate, compete and
exchange
problems with traditional supply chain
management (SCM)

problems this creates for manufacturers

problems this creates for their suppliers

problems this creates for consumers
improvements to traditional SCM

the direct-to-customer model

virtual integration with suppliers
Markets
Networks
Firms
Agents
Quiz
? days a box of cereal spends in the supply
chain?
Distorted information causes total
inventory in the pharmaceutical supply
chain to exceed ? days. $? in savings to be
realized.
$ ? wasted because of poor coordination in
the food industry supply chain
$ ? Boeing write-off in 1997 due to supply
chain inefficiencies
Quiz
A box of cereal spends 104 days in the
supply chain
Distorted information causes total
inventory in the pharmaceutical supply
chain to exceed 100 days. $11 billion in
savings to be realized
Poor coordination wasting $ 30 billion
annually in the food industry
$ 2.6 billion Boeing write-off in 1997 due
to supply chain inefficiencies
Defining SCM
SCM is the coordination of material,
information and financial flows
between and among enterprises
participating in the demand fulfillment
process for a product or service.
Spans multiple organizations and
industries
Coordination and integration of flows
essential for the modern enterprise
Gates: Business @ The Speed of Thoug
A digital nervous system is the corporate,
digital equivalent of the human nervous
system, providing a well-integrated flow of
information to the right part of the
organization at the right time. A digital
nervous system consists of the digital
processes that enable a company to
perceive and react to its environment, to
sense competitor challenges and
customer needs, and to organize timely
responses.
Gates: Business @ The Speed of Thoug
A digital nervous system requires a
combination of hardware and software; it's
distinguished from a mere network of
computers by the accuracy, immediacy, and
richness of the information it brings to
knowledge workers and the insight and
collaboration made possible by the
information.
Scott McNealy on Gates’ View
He is right - I would be very nervous if my
systems were based on their platforms and
products!
 Beer Game video
Traditional supply chain obsolescence
Direction of flow of demand
Direction of flow of product
Point of differentiation
Distribution costs
Market mediation costs
Raw Material
vendor
Tier-II
Suppliers
Tier-I
Suppliers
Manufacturers
Distribution
Centers
Retailers
Customer
Zones
The Bullwhip Effect
Upstream amplification of demand variation
Progression of a brushfire to an inferno!
Customer
Retailer
Distributor
Factory
Tier 1 supplier
Equipment
Machine Tools at Bullwhip Tip
50%
-100%
Data from United States, 1961-1991 (GDP, vehicle production, and machine tool orders
% change GDP
% change vehicle production index
% change net new orders machine tool industry
1991
1989
1987
1985
1983
1981
1979
1977
1975
1973
1971
1969
1967
1965
-50%
1963
0%
1961
% Change, year to year
100%
80
70
60
50
40
30
20
10
0
Week
21
19
17
15
13
11
9
7
5
Factory
Distributor
Wholesaler
Retailer
Customer
3
1
Order
The Diaper Supply Chain!
Ripples to tidal waves
Stockpiles and stockouts
Insufficient or excessive capacities
Higher costs
What is the Problem?

The “bullwhip effect” - four key causes
 Demand signal processing
 Currently only order information is shared (not actual sales)
 Need to instead share POS retail data (sell-through data)
 Order batching (retailers only order periodically)
 Infrequent access to demand information
 Order rationing
 retailers order popular items excessively
 Hoarding of scare products (inflate demand order of scarce
product to ensure that you have it on-hand)
 Special Promotions
 Alter the normal pattern of product demand from customer;
so that it’s impossible to understand the “true” demand
Interorganizational Systems: CRP
BIG RETAILER
< 3% stock outs
Warehouse 1
< 14days inventory
P&G
Warehouse 2
Before CRP
Budget
BIG RETAILER
Actual
Warehouse 1
P&G
Warehouse 2
•Volume discounts
•New product promos
•Here and now discounts
•Trade marketing
•Bonuses….
Interorganizational Systems
 Integration of supply chain across
companies
 Degrees of integration: information,
process, property rights
 Increased efficiencies through
1. optimal production/logistics planning
2. lower inventories
3. increased flexibility
4. customer satisfaction
 Oh brave new world, this is
wonderful…But…
The Economist says….
 Look out for proprietary systems with
high specificity Lock-in
 Sharing processes is optimal from
logistics viewpoint, but remember
‘knowledge of time and place’
 Additional information acquired by
one party can reduce bargaining
power of other. Competitive industries
like retailing, grocery and electronics
has demonstrated many examples of
this….
 Excel video
Solutions to Improve Sales Forecasting
 Vendor Managed Inventory
 Collaborative Forecasting and
Replenishment
 Quantity-Flexible contracts
 (as contrasted with rigid
contracts)
 Buyer allowed to make limited changes to
forecast information, which is then shared
with suppliers. Supplier only ships
enough for the “newest” forecast. Why is
this helpful?
Types of Shared Information
Inventory information
Transition to echelon-based inventory systems
Upstream companies can determine when and
what to produce
Downstream companies can improve service
levels with less inventory
The Apple-Fritz Supplier Hub
Fritz manages entire inbound logistics for Apple
Consolidates freight, clears customs, manages the
hub, manages local transportation to Apple
FLEX system
Types of Shared Information
Sales Data
Variance of orders greater than that of
sales
The “bullwhip effect” - four key causes
Demand signal processing
 Move to sharing sell-through data and POS
retail data
Order batching
 Infrequent access to demand information
Order rationing
 Hoarding of scare products
Promotions
Types of Information Sharing
Production/Delivery Schedule
Improves due-date estimation
Expand planning horizons
Other Information Sharing
Performance metrics
Capacity information
Models of Information Sharing
Information Transfer Model
Transfer information to the other who maintains
the database for decision-making
EDI Limitations
Multiple industry-specific standards
Rigid design for transaction processing
Rigid text formats
Batch-oriented
Installation costs
Models of Information Sharing
Third Party Model
Third party collects and maintains supply
chain information
Apple- Fritz alliance, with Fritz as the 3rd
party
Instill corporation (www.instill.com) in the
food services industry
Digital markets (www.digitalmarket.com) in
the electronic parts market
Challenges
Aligning incentives of different partners
Channel Management Example
Trust and cooperation
Confidentiality of shared information
Anti-trust implications, such as possible
price fixing behavior
Timeliness and accuracy of information
Technological constraints
Commercial Uses of New
Technologies
 Many
commercial
forms are
products of
modern
technologies
Manugistics, I2, Commerce 1, Ariba
3.1
.7
.3
2.3
eCommerce Status? Doing fine….
800
700
600
500
400
300
200
100
0
Billion USD
1999
2,000
2001
eCommerce - Where?
 Manufacturing (18% of all sales)
 Transportation equipment
 Beverage and tobacco
 Electrical equipment & components
 Wholesalers (8% of all sales)
 Drugs and druggists
 Motor vehicles, parts and supplies
 Professional and commercial equipment
eCommerce - Where? (cont.)
 Services (1% of total sales)
 Travel arrangement and reservations
 Securities & commodities
intermediation
 Publishing and software
 Retail Sales (1% of total retail
sales)
 Books and magazines
Outlook




Manufacturing
Wholesalers
Services
Retail Sales
Present
18%
8%
1%
1%
Common
Sense
70%
50%
20%
30%
• 60-80% of all eCommerce conducted through EDI
• x12 & EDIFACT (primarily VANS)
www.census.gov/estats
B2B What Happened ?
 Estimates that over 1,000 B2B
portal will soon consolidate to <
200.
 Less than 15% of all exchanges
operating
 2 Stories:
 Vertical
 Horizontal
Your task….
 You would like to buy a 3 year old
Honda Prelude. You have 2 options:
1. Buy the car in a private transaction,
mediated through the newspaper
classifieds, or
2. Buy the car through a used car dealership
Asses the relative advantages and
disadvantages of each option.
Intermediaries
 Up to 25% of the
economy
 Financial
Intermediaries
 Dealers &
Wholesalers
What do Intermediaries do?
•Infomediation or Information
matching
•Gathering, sorting and arranging
information about buyers and
producers plans in order to match
them.
Intermediation
In addition to information matching, intermediaries also
do:
•Economic matching, in which the intermediaries do
not have the ability to perfectly match producers’ and
consumers’ plans. Therefore they are required to hold
inventories and perform logistic functions where
perfect matching is impossible. The are therefore risk
bearing partners in exchanges. They may also extend
liquidity (credit).
Dis-Intermediation; can it be done?
If both producers and consumers had perfect information, then
they could simultaneously match (infomediate) their plans
and transform intermediaries in to useless entities. (Disintermediation)
Plan matching remains insurmountable, moreover;
1. Information asymmetries are not just about asymmetric
information on plans. Moral hazard and adverse
selection problems remain!!
2. Exogenous forces, uncertainty, bounded rationality
3. There is no logical argument that optimal plans will
match. Many consumption decision are instantaneous
and not planned;exacerbated through liquidity
constraints.
OK, so what do Intermediaries do?
Information management: compiling and filtering
information, informing consumer's knowledge of supply
and demand capacity.
Logistics management: economies of scale, scope and
specialization in conveying goods from production sites
to consumption sites
Transaction securitization: controlling and guaranteeing
the quality of goods and payments delivered to buyer
and seller
Insurance: insurance for the existence of a market for
the products, that is, a market making function
Liquidity: extending credit to both sides of the
transaction, alleviating liquidity constraints
Morgan Stanley



“Collaborative Commerce”
Before the Order





Purchase approval and routing
Promotions and campaigns
Financing
Inventory availability
Price negotiation






Order status
Partial Shipments
Backorder information
Substitute products
Order explosion to multiple suppliers
Scheduling of inventory







Warranty and maintenance
Replacement parts
Asset Management
Regulatory Compliance
Returns and incorrect ships
Settlement
Inspection
During Fulfillment
After Delivery
Remember Amazon
Amazon, when it started, it was mostly and
infomediary. When faced by competition by an
actual commercial intermediary, (Barnes & Noble)
it was forced to purchase in larger quantities and
thus hold inventories. Hence, a commercial
intermediary had advantage over infomediary due
to holdings of inventories and established, well
managed logistics channels. For B&N, their
website is an additional display case. Amazon
may be forced to become a retail distributor...as
was the case for many post order firms at the
beginning of the century!!!!
Determinants of Choice in
Intermediated exchange
Figure 1: Determinants of Choice in Intermediated Exchanges

Determinants
of Uncertainty
perceived market
and environmental
uncertainty.
information
asymmetry,
opportunism, moral
hazard, adverse
selection and
perceived and real
volatility of supply
Hypothetical Decision Path B
Insurance
Logistics Management

B
Liquidity Management

A
Hypothetical Decision Path A
Transaction Securitization
Information
Management


B
temporal sensitivity, immediacy and spatial
needs, aversion to search and coordination
costs
A
Determinants
of Preferences

Relationship to Profits
Table 4a.Variables and Relationship to Profitability
Variable
Expected
Identified
Intermediation state
+
Positive
Diversification level
+
None
Age
+
Weak
Internet pure-play strategy
-
None
Publicly listed
+
None
Table 4b. Variables and Relationship to Intermediation State
Variable
Expected
Identified
Freq. of acquisitions
+
Positive
Freq. of joint venture
-
None
Freq. of organic growth
?
None
Centralization Yields…
Economies of scale in transacting (reduced search,
negotiation and implementation costs)
Economies of scope (group or bundled transactions)
Cross subsidization amongst transactors and transacted
goods play off on different buyer's willingness to pay and
supplier's opportunity costs. ...Can increase welfare overall.
Risk pooling: reduce market uncertainty by holding
inventories.
Rearrangement of goods. When there is a mismatch
between the producer's offering and the buyer's
preferences, the intermediary can rearrange them to
improve the match.
Is Bigger Better ?
 Centralization yields: economies of scale and
scope, risk pooling, liquidity, transparency.
 Between 1921 and 1983, 180 different futures
markets existed, average lifetime of less than 12
years.
 Similar patterns with equity markets…
yes – size does matter!
Network externalities - natural monopoly
1 or 2 markets per sector/service
B2B Portals: prospects…
 Recent study shows less than 15% of B2B
exchanges were actively operating markets
 1000 should consolidate to 200
 Highly standardized products will be
centralized
 Place for niche exchanges in esoteric, nonstandard products with many attributes
 Learn from the consolidation in financial
markets
 Look for fallout & consolidation in
intermediate term
B2B Portals – 2 main types





Horizontal
1 product sector –
many industries
Large exchanges
Provide liquidity,
transparency,
aggregate supply &
demand
Require high volume of
transactions, small
commission base
Additional revenue
through value adds like
financing, asset
management,
warrantees




Vertical
One industry –many
products
Limited membership
Eliminate inefficiencies
in specific industry
supply chains
Fewer transactions –
revenue based on
realized savings
Purpose
 Increase understanding of rent generation
models in electronic intermediaries
 Implications of network and product
characteristics
 Evolution of rent accrual mechanisms &
information and relational capabilities
 Comparative case studies: 2 companies, both
founded in Atlanta in 2000, & backed by large
industry incumbents
 Omnexus
 eGatematrix
Omnexus
Omnexus
 Plastics Industry one of world’s largest
 589 billion dollars in revenue
 Employs 1.5 million people





BASF
Bayer
Dow
Dupont
Ticona/Celanese
Omnexus
•Large marketplace, MCBase
•Search on thousands of materials with specific
properties
•Integration with suppliers ERP systems
•Real time inventory and price data
•Submission of RFQs
•Electronic billing and transaction clearing
•Customer Support
Size
# of Firms Annual Revenues
Large
200
>30 million
Medium
2,700
6 million
Small
5,000
<$1 million
Market Share
>50%
30%
<20%
Segmentation of Resins Buyers
Competition
Evolution
eGate Matrix
Food products, supplies,
materials to caterers or
directly to plane
Suppliers/ Distributors/
Manufacturers
Prepared meals and other
supplies to plane
Equipment (galley, culinary,
etc.) hand-offs between flights
Audio and video supplies/
equipment to plane
Equipment (galley, culinary,
etc.) hand-offs between
flights
Service Providers
Airlines
Caterers
Flight schedule information
from airlines
Quality performance feedback
from airlines
Flight schedule information
from airlines
Quality performance feedback
from airlines
Flight schedule
information from airlines
Quality performance
feedback from airlines
Payment from airlines
Mark-up payment from
airlines for services rendered
Service charges from
airlines
Information Flows
Physical Flows
Financial Flows
Inventory and service
availability updates from all
groups
What Happened? Conclusions






Horizontal
Many portals built on
information aggregation
assumption
Barriers to entry low
Too many portals, can’t
generate volume
Suppliers weary of
transparency (stick with
EDI and Fax)
Most sectors can support
1-3 exchanges (max).
Forget commodities and
content - Focus on
payment, logistics, &
value adds….





Vertical
Often very sound
business model
Implementation hard
work
Barriers to adoption:
legal, organizational,
procedural
Slow in the making,
but scale well
Most profitable in
fragmented markets
with customizable
products