Transcript No Slide Title

Supply Chain Issues in
Small Agricultural Enterprises
by
Cerry M. Klein and Wooseung Jang
Department of Industrial and Manufacturing Systems Engineering
Preview
•
•
•
•
Introduction
Problems and Objectives
Models
Conclusions
Department of Industrial and Manufacturing Systems Engineering
Introduction
• Agricultural Production
• Large sector of economy
• Leading exporter
• Embraces new technology
• Derivatives, GPS, etc.
• Small Farm Enterprise
•
•
•
•
350,000 “Large Farms” – 95% of inputs – 95% of outputs
63% decline in farm ownership – 97.5% decline for minorities
1,000 acres needed to break even
Federal Subsidies - $27 billion – 10% to farm owners – most of
which were large farm owners (policies not helping)
Department of Industrial and Manufacturing Systems Engineering
Introduction
• Concentration Ratios in Agriculture
• CR4 = % market controlled by 4 largest firms
• Flour Milling
• CR4 (1982) = 40%; CR4 (1997) = 62%
• Soybean Crushing
• CR4 (1977) = 54%; CR4 (1997) = 80%
• Pork Packing
• CR4 (1987) = 37%; CR4 (1998) = 57%
• Beef Packing
• CR4 (1990) = 72%; CR4 (1998) = 79%
Source: Heffernan (1999)
Department of Industrial and Manufacturing Systems Engineering
Introduction
• The Farming Enterprise
• $185 billion annually on inputs such as chemicals, seeds, land,
supplies, etc. and in return sales of $210 billion worth of outputs
• 1950’s net farm income as a percent of gross farm income was 35
percent. Today it is 17%
• Industrialized agriculture has embraced new technologies and
management procedures – including supply chain management and
web-enabled methodology
• The small farmer does not have the size or production capacity to
realize the necessary efficiencies of industrialized agriculture
Department of Industrial and Manufacturing Systems Engineering
Introduction
220
200
180
Retail Price
160
Index (1982-84 = 100)
Farmers’ Share
of the Food
Dollar, 1954 to
1998
140
Farm Value Share of Retail Price
120
100
80
60
40
20
Source: Economic Research
Service (2000)
0
1954 1958 1962 1966 1970 1974 1978 1982 1986 1990 1994 1998
Department of Industrial and Manufacturing Systems Engineering
Introduction
• Rural Communities and The Farming Enterprise
• “Decentralized land ownership produces more equitable economic
opportunity for people in rural communities, and offers self employment
and business management opportunities. Farms, particularly family farms,
can be nurturing places for children to grow up and acquire the values of
responsibility and hard work.”
• “As small farms and farm-workers succeed...they will fuel local
economies and energize rural communities across America.” The National
Commission on Small Farms (1998)
Department of Industrial and Manufacturing Systems Engineering
Introduction
• Small Farm Enterprises
• For the small farm enterprise to be viable it must be able to
respond quickly to product differentiation and to establish niche
areas of product
“farmers will increasingly be producing commodities with specific attributes called for by food
processors who are responding to retail demand. Traditional patterns of farming will change; more
products will be produced for niche markets and for international tastes. There will be higher pay-off
for the entrepreneur on the farm” (Kinsey and Senaur)
“market segregation may provide niche opportunities for producers who are willing to keep their
product segregated and sell based on specific attributes rather than in bulk.” (Baumel)
Department of Industrial and Manufacturing Systems Engineering
Introduction
• Small Farm Enterprises
• “Rural people are disadvantaged regarding access to the basic
technical knowledge to use the expanding infrastructure
effectively”. U.S. Dept. of Commerce report on the digital divide
(U.S.DoC, 1999)
• This is borne out in the Hopkins and Morehart (2001) study that
shows only 0.33 percent of all sales and purchases in agriculture
during 2000 were conducted online. In addition only 1% of all
farms connected to the internet conducted sales online.
Department of Industrial and Manufacturing Systems Engineering
Introduction
• Small Farm Enterprises
• Do not have readily available tools and methodology to assist in
effectively accessing or “organizing” new value-added or niche
markets for their products
• They lack the capability to take advantage of the newly developed
technologies in information systems to construct supply/value
chains that reduce their vulnerability to risk while increasing their
direct profit.
• An area devoid of contributions at the research level
Department of Industrial and Manufacturing Systems Engineering
Problems and Objectives
• Strategic Decision Making
• “links between food manufacturers, wholesalers, and retailers are
complex, ill-understood, and changing rapidly” Kinsey (2000)
• Logistics
• unlikely that a group of homogeneous small farmers will be in a common
geographical area
• Information Technology and E-commerce
•
•
•
web-based technologies and e-commerce will be essential tools in the development
and management of supply chains
The advantage of this is to provide enhanced market access for rural small business
and for organizing markets and supply chains
“the role of information technology has largely been ignored” (Buhr 2000)
Department of Industrial and Manufacturing Systems Engineering
Problems and Objectives
To efficiently deliver products small farmers need to:
• Create or reengineer their supply operations to meet the
requirement of speed and flexibility
• Integrate and coordinate the system, which includes
• customers, suppliers, information, productions, inventories,
transportations, quality, prices, partnerships, and interdependencies.
• The lack of predictive models to analyze supply chains and ecommerce might be part of the reason for the low visibility and
chaotic situations that exist in supply operations.
• New business models are needed to help assimilate, simplify and
manage supply chain operations.
Department of Industrial and Manufacturing Systems Engineering
Problems and Objectives
To become and remain competitive through niche
markets small farmers need to:
• Develop virtual co-ops for leverage
• Embrace e-commerce and web-based technologies to
expand their markets.
• Interact globally with suppliers and buyers
• Create “new economic spaces” (Greenspan, Kelly and
Gottlieb and Fisher)
Department of Industrial and Manufacturing Systems Engineering
Models
Strategic
Form Co-ops
Strategic
Tactical
Contract/Pricing
Size of Co-op
E-Commerce
Acres to Plant
Strategic
Operational
Shipment Mode
Planting
Amounts of
Pesticides, Fert.
Weather
Phenomena
Irrigation
Inv entory
Product
Selection
Scheduling
Distribution
Pesticides
T ransport
Harvesting
Fertilizers
Holding
T ime
Figure 1. Samples of Agricultural Enterprise Decisions
Department of Industrial and Manufacturing Systems Engineering
Models
• Strategic Planning and Decision Making
• Decisions include:
•
•
•
•
To join or form a co-op
What type of product to produce and how much
Determination of when to take a product to the market
Contract and pricing, capacity decisions, etc.
• To join or form a co-op seems to be a very
important decision and is considered here.
Department of Industrial and Manufacturing Systems Engineering
Models
• Co-op Problem
• Without joining co-op can only sell individually and
directly to customer
• If join a co-op, usually more stable demand and larger
customer base. Also, not necessary to sell directly to
customer
• Question is whether to join or not, which is dependent
on co-op size. What is the optimal size of the co-op?
Department of Industrial and Manufacturing Systems Engineering
Models
• Co-op Problem
• Assumptions
• Multiple homogeneous farmers producing one specific product
• Production quantity is assumed to be Q
• Each farmer faces local demand X, which is stochastic and has
probability density function  (x ) and cumulative distribution
function  (x )
• The price of the product for the local and direct farmer-consumer
sales is equal to p1
• local demand and price is not affected by outside factors such as other
co-ops and grocery stores
Department of Industrial and Manufacturing Systems Engineering
Models
• Co-op Problem
• Suppose that the demand from wholesalers and institutions is
expected to be D, and the sales price per item is , where p2<p1
• If n farmers form a co-op, total production quantity is nQ > D.
Assume each farmer sells D/n items through the co-op and the rest,
Q-D/n, to local customers
• If a farmer only sells products directly to local customers without
using a co-op, then the revenue given by is f(0)
Q

0
Q
f (0)   p1 x ( x)dx   p1Q ( x)dx
Department of Industrial and Manufacturing Systems Engineering
Models
If a farmer is a member of an n-farmer co-op, then the revenue is given by
f (n)  p2 D / n  
QD / n
0
p1 x ( x)dx  

QD / n
p1 (Q  D / n) ( x)dx
It is possible to show that f(n) is an increasing function in terms of n  1 , when
Q   1 (1  p2 / p1 )
and unimodal if
Q  1 (1  p2 / p1 )
Department of Industrial and Manufacturing Systems Engineering
Models
Can use this model to determine whether or not to join a co-op.
There exists a Q* and an n* such that
If Q  1 (1  p2 / p1 )
then
f (0)  f (n) n
If  (1  p2 / p1 )  Q  Q *
 f (0)  f (n)
then 
 f (0)  f (n)
If Q  Q *
then
1
n  n*
n  n*
f (0)  f (n) n
Department of Industrial and Manufacturing Systems Engineering
Models
• In other words:
• A farmer should not join a co-op if the expected marginal price of
direct selling is larger than the price obtained through the co-op
participation
• A farmer should always join a co-op if his production quantity is
larger than a certain threshold value
• Otherwise, the farmer’s decision should depend on the size of a coop.
• The values of Q* and n* can be explicitly computed and used to
assist individual farmers who try to measure the value of forming
and/or joining a co-op.
Department of Industrial and Manufacturing Systems Engineering
Models
• B2B and B2C Models
• When are direct sales to customers profitable for the small
farmer?
• Quantity and pricing strategies to optimize farmers’ profits
• Because many products are perishable, farmers need to find
balances between lower, yet certain profits, and higher
profits with uncertainties
• Analytical models can provide operational decisions so that
farmers’ overall expected profits are maximized
Department of Industrial and Manufacturing Systems Engineering
Models
• B2B and B2C Models
• Once a co-op is formed, various operational decisions should be
made. Both the co-op and wholesalers desire to make decisions
that maximize their profits, but only coordinated decisions can
realize this objective
• Joint coordination can result in decreased operation costs and
reduced uncertainties, which will eventually yield greater
satisfaction for participants. This is widely done in industrialized
agriculture but is virtually nonexistent in small farm operations,
(Hobbs and Young 2000)
Department of Industrial and Manufacturing Systems Engineering
Models
• B2B and B2C Models
• Example: Supply Contract and Pricing Decisions
• Assume:
• Co-op suggests the sales price (p2) for its products to wholesalers
• Wholesaler decides on an order quantity (D) so as to minimize
the total cost
• Wholesaler faces random demand later from retailers and may
need to pay either a holding cost (h) or a shortage penalty (s)
• The pdf and cdf or the random demand is given by:
 (x )
and
 (x )
Department of Industrial and Manufacturing Systems Engineering
Models
• B2B and B2C Models
• The profit function for the supply chain is equal to the
revenue of the wholesaler minus the wholesaler’s
holding and shortage costs minus the co-op’s
production cost

D

0
0
D
f ( D)  p  x ( x)dx  h ( D  x) ( x)dx  s  ( x  D) ( x)dx  v( D)
Where  is the sales price of the wholesaler
Department of Industrial and Manufacturing Systems Engineering
Models
• B2B and B2C Models
• If the wholesaler follows an optimal ordering policy
based on the well-known Newsboy problem, then
 s  p2 
D   1 

 hs 
and the profit function can be rewritten as follows
 s  p2 
 1 

 h s 
0
 1  s  p2  
  
  x  ( x)dx
0
h

s
 
 


 1  s  p2  
 s  p2  
 s  1  s  p2   x   1 


(
x
)
dx

v
  

 
 

 h  s 
  h  s 
 h s  
_

f ( p2 )  p  x ( x)dx  h 
Department of Industrial and Manufacturing Systems Engineering
Models
• B2B and B2C Models
• One of the objectives is to decide an optimal transaction
price between the co-op and the wholesaler so that
above function is maximized. After some analysis, we
can show that the optimal transaction price satisfies
s  p2
 ( v ' ) ( p2 )  
hs
1
Department of Industrial and Manufacturing Systems Engineering
Models
• B2B and B2C Models
• This results show that the optimal price is decreasing in demand if the
production cost is strictly concave while the price is constant if the
production cost is linear.
• This implies, if a product has a concave production cost, a larger co-op
can offer the product at a lower price enticing more wholesalers to the
supply chain, and the addition will provide a profit increase for all existing
members.
• We can conjecture that eventually only well-coordinated large co-ops will
survive. Therefore, it is necessary for a co-op to be aggressive in
increasing demand, especially at the beginning stage of the supply chain.
• If production cost is linear then smaller co-ops can compete because the
optimal transaction price should be the same regardless of the size. Thus,
the operation of a co-op should focus on other issues such as quality and
customer service rather than becoming large.
Department of Industrial and Manufacturing Systems Engineering
Models
• Accepting a New Contract
• Contract between buyer and cooperative defined by
price and quantity (p,D)
• Let p0 be price for local buyer now
• Assume have current contract (p1,D1) and considering
new contract (p2, D2) where nQ>D1+D2
• Accept New Contract if
n
p2  p0 (1 
D2

Q  D1 n
Q ( D1  D2 ) n
( x)dx))
Department of Industrial and Manufacturing Systems Engineering
Models
• Amount of Product for Farmer to Supply
• Suppose that the farmer contributes R products then the revenue is
g ( R)  p0Q  ( p0  p1 ) R  p0 
Q R
0
( x)dx
• The optimal delivery quantity for the farmer assuming they are
required to deliver a quantity between R1and R2, R1 < R2 is then
Q
R1
Q -  -1(1- p1 /p0)
R2
if Q<R1
if R1 < Q < R1 +  -1(1-p1 /p0)
if R1 +  -1(1- p1 /p0) < Q < R2 + -1(1- p1 /p0)
otherwise
Department of Industrial and Manufacturing Systems Engineering
Conclusions
• Initial stages of study
• Need real data
• Co-ops in Mississippi, California, Illinois, and Missouri
•
•
•
•
Models need to be extended to more realistic scenarios
Problems difficult, but interesting due to the uncertainties
Important problem from a political and policy standpoint
Supply chains are in a highly variable, high risk, low
margin, seasonal, niche market environments – may relate
or be applied to certain “dot.coms”
Department of Industrial and Manufacturing Systems Engineering