Bullwhip Effect: Demand

Amplification

Lecture by:
Prof M. K. Tiwari
Department of Industrial Engineering and
Management
Indian Institute of Technology, Kharagpur
Bullwhip Effect Defined
The bullwhip effect is the uncertainty
caused from distorted information
flowing up and down the supply chain.
Results of the bullwhip effect
 Excess inventories
 Problems with quality
 Increased raw material costs
 Overtime expenses
 Increased shipping costs
Results of the bullwhip effect -
continued…
 Lost customer service
 Lengthened lead time
 Lost sales
 Unnecessary adjusted capacity
Causes of “Bullwhip” Effect

1 Demand Forecast Updating

2 Order Batching

3 Price Fluctuation

4 Rationing and Shortage Gaming

 Demand Forecast Updating

 Orders as signals of product demand

 Forecasting techniques that rely heavily on

recent demand observations

 Safety Stocks that depend on erratic variance

of demand and long leadtimes

 Long leadtimes and forecasting difficulties

 Order Batching

 Economies of scale in ordering costs and

manufacturing setups lead to batching

 Periodic planning and ordering as part of MRP

and periodic review systems

 Economies of scale in transportation (full

truckload rates)

 “Push” ordering motivated by short term

financial performance measures
  Price Fluctuation
 Quantity discounts and advance purchases
 Variance of buying quantities bigger than variances in consumption
rate
Bullwhip Effect due to Seasonal Sales of Soup
800 Shipments from Manufacturer
Order Quantity

600 to Distributors
400 Retailers’ Sales
200
0
1 Weeks 52

 Wide swings result in capacity adjustment costs, high inventory costs,

transport diseconomies
 Trade promotions and supply chain management = “The dumbest
marketing ploy ever!”
 Rationing and Shortage
Gaming
 When demand exceeds manufacturing
capacity, rationing or orders
 Customers exaggerate their real needs
(“gaming”)
 When capacity constraints are removed,
orders suddenly drop
Other causes of the bullwhip
effect
 Un-forecasted sales promotions
 Sales incentives
 Lack of customer confidence
 Customers turning back sales orders
 Freight incentives
Bullwhip effect - an example

Chronology of company “X’s” supply chain

problem.
 Company X produces widgets for sale
on the open market.
 Customer demand for Company X’s
widgets become stagnant
 Retailers offer a sales promotion to
boost sales of Company X widgets
Example – continued
 Retailers fail to notify manufacturers of
sales promotion
 Company X recognizes that demand
for widgets has increased.
 Company X increases inventory to
allow for increased manufacturing of
widgets.
Example - continued
 Company X notifies part suppliers of
increased demand.
 Suppliers increase inventory to meet
demand.
 Increasing Variability of Orders up the
Supply Chain

Consumer Sales Retailer’s Order to Manufacture

20 20
Order Quantity

Order Quantity
15 15
10 10
5 5
0 0
Time Time

Wholesaler’s Orders to Manufacture Manufacturer’s Orders to Supplier

20 20
Order Quantity

Order Quantity
15 15
10 10
5 5
0 0
Time Time
Solving the Bullwhip dilemma
 Improve communication along the supply
chain.
 Retailers notifying firms upstream of sales
promotions will help clarify demand signals
from consumers
 Improved information will improve demand
forecasts upstream in the supply chain.
Solving the Bullwhip dilemma -
continued
 Improve sources of forecast data
Firms can use data from Point of Sale
computer systems to derive data from
forecasting
Firms along the supply chain can use EDI
systems to retrieve data on items that are
legitimately being purchased by customers
Solving the Bullwhip dilemma -
continued

 Work with firms upstream and

downstream in the supply chain
 Create smaller order increments to decrease
time between orders. Order processing will
become closer to real-time.
 Work to develop consistent pricing of
products to avoid demand fluctuations from the
sale of inexpensive products.
How to counteract the
“Bullwhip” Effect

A Avoid Multiple Demand Forecast

Updates

B Break Order Batches

C Stabilize Prices

D Eliminate Gaming in Shortage Situations

 A. Framework for supply chain
coordination initiatives

Cause of  Demand forecast update

Bullwhip:

 Understanding System Dynamics

Information  Use of Point-of-Sale (POS) data
Sharing:  Electronic Data Interchange (EDI)
 Internet
 Computer Assisted Ordering
 The Bullwhip Effect
and its Impact on the Supply Chain

 Consider the order pattern of a single color television model sold by a

large electronics manufacturer to one of its accounts, a national retailer.
The Bullwhip Effect
and its Impact on the Supply Chain

Point-of-sales Data-
Original

POS Data After

Removing
Promotions
The Bullwhip Effect
and its Impact on the Supply Chain

POS Data After Removing Promotion & Trend

We Conclude ….

 Order variability is amplified up

the supply chain; upstream
echelons face higher variability.
 What you see is not what they
face.
What are the Causes….
 Promotional sales
 Volume and transportation discounts
 Inflated orders
 Demand forecasting
• Order-up-to points are modified as forecasts
change – orders increase more than forecasts
 Long cycle times
• Long lead times magnify this effect
What are the Causes….
 Promotional sales
• Forward buying
 Volume and transportation discounts
• Batching
 Inflated orders
• IBM Aptiva orders increased by 2-3 times when
retailers thought that IBM would be out of stock over
Christmas
• Motorola cell phones
What are the Causes….

 Single retailer, single manufacturer.

• Retailer observes customer demand,
• Retailer orders qt from manufacturer.

Dt qt
Retailer Manufacturer
L
 How big is the increase?

 Suppose a P period moving average is used.

2
Var(q) 2L 2L
 1  2
Var( D) P P
Var(q)/Var(D):
For Various Lead Times

14
L=5
12

10

8 L=3
6

4 L=1
L=1
2

0
0 5 10 15 20 25 30
Consequences….

 Increased safety stock

 Reduced service level

 Inefficient allocation of
resources
 Increased transportation
costs
 A. Framework for supply chain
coordination initiatives (cont.)

 Vendor Managed Inventory

(VMI)
Channel  Continuous Replenishment
Alignment: Program (CRP) at Campbell
Soup, Nestle, M&M, P&G, Scott
Paper
 Consumer Direct Programs
 Discount for Information Sharing
 Lead time reduction
Operational
 Echelon-based inventory control
Efficiency:
 B. Framework for supply chain
coordination initiatives

Cause of  Order Batching

Bullwhip:

 EDI
Information
 Internet Ordering (Trading Process
Sharing:
Network at GE)
 B. Framework for supply chain
coordination initiatives (cont.)

 Discount for Truck Load Assortment

Channel  Delivery appointments
Alignment:  Consolidation
 Mixed SKUs
 Third Party Logistics

Operational  Reduction in fixed costs of ordering by

Efficiency: EDI or electronic commerce
 CAO
 C. Framework for supply chain
coordination initiatives

Cause of
Bullwhip:  Price Fluctuations

 Continuous Replenishment Program

Information
(CRP)
Sharing:
 Everyday Low Cost (EDLC)

Operational  Everyday Low Price (EDLP)

Efficiency:  Activity Based Costing (ABC)
 D. Framework for supply chain
coordination initiatives
Cause of
Bullwhip:
 Shortage Gaming
Information
Sharing:  Sharing Sales, capacity and inventory
data

Channel
Alignment:  Allocation based on past sales
 Placing orders ahead of peak season
 Stricter return and order cancellation
policies
Countermeasures to the
Bullwhip Effect
Countermeasures to the Bullwhip Effect

1. Countermeasures to order batching

2. Countermeasures to shortage gaming
3. Countermeasures to fluctuating prices
4. Countermeasures to demand forecast
inaccuracies
5. Free return policies
Order Batching

 High order cost is countered with Electronic Data

Interchange (EDI) and computer aided ordering (CAO).
• Full truck load economics are countered with third-party
logistics and assorted truckloads. Random or correlated
ordering is countered with regular delivery appointments.
• More frequent ordering results in smaller orders and smaller
variance.
• However, when an entity orders more often, it will not see
a reduction in its own demand variance - the reduction is
seen by the upstream entities.
• Also, when an entity orders more frequently, its required
safety stock may increase or decrease; see the standard
loss function in the Inventory Management section.
Shortage Gaming
 Proportional rationing schemes are countered
by allocating units based on past sales.
• Ignorance of supply chain conditions can be
addressed by sharing capacity and supply information.
• Unrestricted ordering capability can be addressed by
reducing the order size flexibility and implementing
capacity reservations.
• For example, one can reserve a fixed quantity for a
given year and specify the quantity of each order
shortly before it is needed, as long as the sum of the
order quantities equals to the reserved quantity.
Fluctuating Prices
 High-low pricing can be replaced with
every day low prices (EDLP). Special
purchase contracts can be implemented
in order to specify ordering at regular
intervals to better synchronize delivery
and purchase.
Demand Forecast Inaccuracies
 Lack of demand visibility can be addressed by
providing access to point of sale (POS) data.
 Changes in pricing and trade promotions and
channel initiatives, such as vendor managed
inventory (VMI), coordinated forecasting and
replenishment (CFAR), and continuous
replenishment can significantly reduce demand
variance.
Free Return Policies
 Free return policies are not addressed
easily.
 Often, such policies simply must be
prohibited or limited.
Vendor Managed
Inventory
 Popularized in the late 1980s by Wal-Mart and
Procter & Gamble, VMI became one of the key
programs in the grocery industry’s pursuit of
“efficient consumer response” and the garment
industry’s “quick response.”
 Successful VMI initiatives have been
trumpeted by other companies in the United
States, including Campbell Soup and Johnson
& Johnson, and by European firms like Barilla
(the pasta manufacturer).
The VMI Partnership
 The supplier—usually the manufacturer but sometimes a
reseller or distributor—makes the main inventory
replenishment decisions for the consuming organization.
• The supplier monitors the buyer’s inventory levels
(physically or via electronic messaging) and makes periodic
resupply decisions regarding order quantities, shipping, and
timing.
• Transactions customarily initiated by the buyer (like
purchase orders) are initiated by the supplier instead.
• The purchase order acknowledgment from the supplier may
be the first indication that a transaction is taking place; an
advance shipping notice informs the buyer of materials in
transit.
The
manufacturer is
responsible for
both its own
inventory and
the inventory
stored at is
customers’
distribution
centers.
Demand Amplification
in Supply Chain
 Multi-Stage Supply Chains
 Consider a multi-stage supply chain:
• Stage i places order qi to stage i+1.
• Li is lead time between stage i and i+1.

qo=D Retailer q1 Manufacturer q2 Supplier

Stage 1 L1 Stage 2 Stage 3
L2
 Multi stage systems

 Centralized: each stage bases orders on retailer’s

forecast demand. 2
 k
 k

k 2 L 2  L  i i
Var (q ) 
 1 
i 1
 i 1

Var ( D) P P2

 Decentralized: each stage bases orders on previous

stage’s demand
k 
2 Li 2 Li 
2
Var (q k )
  1   2 
Var ( D) i 1  P P 
Multi-Stage
Systems:Var(qk)/Var(D)

30
Dec, k=5
25
20
15
Cen, k=5
10
Dec, k=3
5 Cen, k=3
k=1
0
0 5 10 15 20 25
The Bullwhip Effect:
Managerial Insights
 Exists, in part, due to the retailer’s need to
estimate the mean and variance of demand.
 The increase in variability is an increasing
function of the lead time.
 The more complicated the demand models and
the forecasting techniques, the greater the
increase.
 Centralized demand information can
significantly reduce the bullwhip effect, but will
not eliminate it.
Example:
Quick Response at Benetton

 Benetton, the Italian sportswear manufacturer,

was founded in 1964. In 1975 Benetton had
200 stores across Italy.
 Ten years later, the company expanded to the
U.S., Japan and Eastern Europe. Sales in
1991 reached 2 trillion.
 Many attribute Benetton’s success to
successful use of communication and
information technologies.
Example:
Quick Response at Benetton

 Benetton uses an effective strategy, referred to

as Quick Response, in which manufacturing,
warehousing, sales and retailers are linked
together. In this strategy a Benetton retailer
reorders a product through a direct link with
Benetton’s mainframe computer in Italy.
 Using this strategy, Benetton is capable of
shipping a new order in only four weeks,
several week earlier than most of its
competitors.
How Does Benetton
Cope with the Bullwhip Effect?

1. Integrated Information Systems

• Global EDI network that links agents with
production
and inventory information
• EDI order transmission to HQ
• EDI linkage with air carriers
• Data linked to manufacturing
2. Coordinated Planning
• Frequent review allows fast reaction
• Integrated distribution strategy
Information for Effective
Forecasts
 Pricing, promotion, new products
• Different parties have this information
• Retailers may set pricing or promotion without
telling distributor
• Distributor/Manufacturer might have new
product or availability information
 Collaborative Forecasting addresses
these issues.
Information for Coordination of
Systems
 Information is required to move from local to global
optimization
 Questions:
• Who will optimize?
• How will savings be split?
 Information is needed :
• Production status and costs
• Transportation availability and costs
• Inventory information
• Capacity information
• Demand information
Locating Desired Products
 How can demand be met if products are
not in inventory?
• Locating products at other stores
• What about at other dealers?
 What level of customer service will be
perceived?
Lead-Time Reduction
 Why?
• Customer orders are filled quickly
• Bullwhip effect is reduced
• Forecasts are more accurate
• Inventory levels are reduced
 How?
• EDI
• POS data leading to anticipating incoming orders.
Information to Address Conflicts
 Lot Size – Inventory:
• Advanced manufacturing systems
• POS data for advance warnings
 Inventory -- Transportation:
• Lead time reduction for batching
• Information systems for combining shipments
• Cross docking
• Advanced DSS
 Lead Time – Transportation:
• Lower transportation costs
• Improved forecasting
• Lower order lead times
 Product Variety – Inventory:
• Delayed differentiation
 Cost – Customer Service:
• Transshipment
 The Effect of Lack of
Coordination on Performance
 Manufacturing cost (increases)
 Inventory cost (increases)
 Replenishment lead time (increases)
 Transportation cost (increases)
 Labor cost for shipping and receiving (increases)
 Level of product availability (decreases)
 Relationships across the supply chain (worsens)
 Profitability (decreases)
 The bullwhip effect reduces supply chain profitability by
making it more expensive to provide a given level of product
availability
Obstacles to Coordination
in a Supply Chain
 Incentive Obstacles
 Information Processing Obstacles
 Operational Obstacles
 Pricing Obstacles
 Behavioral Obstacles
Incentive Obstacles
 When incentives offered to different stages or
participants in a supply chain lead to actions that
increase variability and reduce total supply chain
profits – misalignment of total supply chain objectives
and individual objectives
 Local optimization within functions or stages of a
supply chain
 Sales force incentives
Information Processing
Obstacles
 When demand information is distorted as it moves
between different stages of the supply chain, leading
to increased variability in orders within the supply
chain
 Forecasting based on orders, not customer demand
• Forecasting demand based on orders
magnifies demand fluctuations moving up the
supply chain from retailer to manufacturer
 Lack of information sharing
Operational Obstacles
 Actions taken in the course of placing and filling
orders that lead to an increase in variability
 Ordering in large lots (much larger than dictated by
demand)
 Large replenishment lead times
 Rationing and shortage gaming (common in the
computer industry because of periodic cycles of
component shortages and surpluses)
Pricing Obstacles

 When pricing policies for a product

lead to an increase in variability of
orders placed
 Lot-size based quantity decisions
 Price fluctuations (resulting in forward
buying)
 Behavioral Obstacles

 Problems in learning, often related to communication in the supply

chain and how the supply chain is structured
 Each stage of the supply chain views its actions locally and is unable
to see the impact of its actions on other stages
 Different stages react to the current local situation rather than trying
to identify the root causes
 Based on local analysis, different stages blame each other for the
fluctuations, with successive stages becoming enemies rather than
partners
 No stage learns from its actions over time because the most
significant consequences of the actions of any one stage occur
elsewhere, resulting in a vicious cycle of actions and blame
 Lack of trust results in opportunism, duplication of effort, and lack of
information sharing
Managerial Levers to
Achieve Coordination
 Aligning Goals and Incentives
 Improving Information Accuracy
 Improving Operational Performance
 Designing Pricing Strategies to
Stabilize Orders
 Building Strategic Partnerships and
Trust
Aligning Goals and Incentives
 Align incentives so that each participant
has an incentive to do the things that will
maximize total supply chain profits
 Align incentives across functions
 Pricing for coordination
 Alter sales force incentives from sell-in (to
the retailer) to sell-through (by the retailer)
Improving Information Accuracy

 Sharing point of sale data

 Collaborative forecasting and
planning
 Single stage control of replenishment
• Continuous replenishment programs
(CRP)
• Vendor managed inventory (VMI)
Improving Operational Performance

 Reducing replenishment lead time

• Reduces uncertainty in demand
• EDI is useful
 Reducing lot sizes
• Computer-assisted ordering, B2B exchanges
• Shipping in LTL sizes by combining shipments
• Technology and other methods to simplify receiving
• Changing customer ordering behavior
 Rationing based on past sales and sharing information to limit
gaming
• “Turn-and-earn”
• Information sharing
 Designing Pricing Strategies
to Stabilize Orders
 Encouraging retailers to order in smaller lots and reduce
forward buying
 Moving from lot size-based to volume-based quantity
discounts (consider total purchases over a specified time
period)
 Stabilizing pricing
• Eliminate promotions (everyday low pricing, EDLP)
• Limit quantity purchased during a promotion
• Tie promotion payments to sell-through rather than amount
purchased
 Building strategic partnerships and trust – easier to implement
these approaches if there is trust
Building Strategic Partnerships
and Trust in a Supply Chain
 Background
 Designing a Relationship with
Cooperation and Trust
 Managing Supply Chain Relationships
for Cooperation and Trust
Building Strategic Partnerships
and Trust in a Supply Chain
 Trust-based relationship
• Dependability
• Leap of faith
 Cooperation and trust work because:
• Alignment of incentives and goals
• Actions to achieve coordination are easier to
implement
• Supply chain productivity improves by reducing
duplication or allocation of effort to appropriate
stage
• Greater information sharing results
Trust in the Supply Chain
 Table 17.2 shows benefits
 Historically, supply chain relationships are based
on power or trust
 Disadvantages of power-based relationship:
• Results in one stage maximizing profits, often at
the expense of other stages
• Can hurt a company when balance of power
changes
• Less powerful stages have sought ways to resist
Building Trust into a
Supply Chain Relationship
 Deterrence-based view
• Use formal contracts
• Parties behave in trusting manner out of self-
interest
 Process-based view
• Trust and cooperation are built up over time as a
result of a series of interactions
• Positive interactions strengthen the belief in
cooperation of other party
 Neither view holds exclusively in all situations
Building Trust into a
Supply Chain Relationship
 Initially more reliance on deterrence-based
view, then evolves to a process-based view
 Co-identification: ideal goal
 Two phases to a supply chain relationship
• Design phase
• Management phase
Designing a Relationship
with Cooperation and Trust
 Assessing the value of the
relationship and its contributions
 Identifying operational roles and
decision rights for each party
 Creating effective contracts
 Designing effective conflict resolution
mechanisms
Assessing the Value of the Relationship
and its Contributions

 Identify the mutual benefit provided

 Identify the criteria used to evaluate the
relationship (equity is important)
 Important to share benefits equitably
 Clarify contribution of each party and the
benefits each party will receive
Creating Effective Contracts
 Create contracts that encourage
negotiation when unplanned contingencies
arise
 It is impossible to define and plan for every
possible occurrence
 Informal relationships and agreements can
fill in the “gaps” in contracts
 Informal arrangements may eventually be
formalized in later contracts
Designing Effective Conflict
Resolution Mechanisms

 Initial formal specification of rules and

guidelines for procedures and
transactions
 Regular, frequent meetings to
promote communication
 Courts or other intermediaries
Managing Supply Chain Relationships for
Cooperation and Trust

 Effective management of a relationship is

important for its success
 Top management is often involved in the
design but not management of a
relationship
 Figure 17.5 -- process of alliance evolution
 Perceptions of reduced benefits or
opportunistic actions can significantly
impair a supply chain partnership
Achieving Coordination in Practice
 Quantify the bullwhip effect
 Get top management commitment for coordination
 Devote resources to coordination
 Focus on communication with other stages
 Try to achieve coordination in the entire supply
chain network
 Use technology to improve connectivity in the
supply chain
 Share the benefits of coordination equitably
 Price fluctuation
When, product’s price is low (through the direct discount or any
promotional scheme), customer buys in bigger amount than the
needed. When product’s price returns to normal the customer
stops buying until it has deleted its inventory. Thus, High-low
pricing produces the difficulty to access the actual demand of
product.

 Contributing factors
• High-Low Pricing leading to forward buy
• Delivery and Purchase not synchronized
 Counter Measures
• Every day low prices (EDLP)
• Limited purchase quantities
• Scan based promotions
Our Focus!
 Our Focus is on the counter measure of bullwhip effect,
where a policy based replenishments program are
developed to establish coordination between each
participated stages of supply chain.
 Considers the situation
• The participants doesn’t share their demand information with other
members.
• Each player has more than one choices of transportations (fast
transportation facility, slow transportation facility).
• The participants follows the same replenishment policy that is
assigned by manufacturer to all of the participants.
Continue…
Optimal order up-to level based policy: According to this policy, the
optimal order up-to level (s) are assigned to the each participants of
the supply chain and they utilize it to decide the amount of product
that is to be ordered for replenishment.

An example:
• It is a multi-stage, serial and single product supply chain
• Supply chain comprises (Customer–Retailer–Warehouse-Distributor
and Manufacture)
• Each stage participants orders to immediate up-stream member of the
supply chain
• Manufacturer has database of the past demands ( Data of per period
customer demand )
• Retailer , warehouse and distributor have two types of transportation
options (their lead times are one period and two periods respectively)
to effectively replenish their demands.
Approach for replenishment
Traditional order up-to level approach- According to this approach
each member forecast their order up-to level in each period and
creates orders to achieve their inventory level up-to the forecasted
level.

Optimal order up-to level approach- In this approach, a fixed order

up-to level (s) is assigned to each supply chain participants by a
governing body (Manufacturer). All member follows this assigned
optimal level for long period of time until a new order up-to levels
are allotted.
 Functioning of traditional order up-to
level approach
Lead time Lead time

order order
Customer demand
Retailer Distributor Manufacture

Order Order
replenishment replenishment

• Forecast for next

period customer • Forecast for next • Forecast for next
demand period retailers period distributors
• Current inventory demand demand
level • Current inventory • Current inventory
level level
 Process performed by retailer at a
period t
Yt= lead time* ( moving average
of demand in time period P) +
adjusting constant *(variance
of lead time demand) (qt) Order created at the
End of period t -1

Retailer forecast Order (qt)=

order upto point order upto point (yt)- remaining
Demand at for time t (yt) Unit at end of t-1(yt-1-Dt-1)
Time t
Retailer

Replenishment
of order at
starting of time t
Remaining unit at
End of (t-1) order-up-to
inventory level at
time(t-1) – demand
realized by retailer
at (t-1)
 Optimal order up-to level
approach
Optimal order up-to level approach-
It establish a continuous replenishment program. In the starting of the
period participant realize their inventory level and the amount it get below
to the assigned order up-to inventory level, they order to immediate
upstream member to replenish it.
Manipulation of optimal order up-to levels:
Objective-
• Maximization of acquired profit
• Maximum customer satisfaction
• Minimization of supply line inventory
• Minimum transportation cost
 Manipulation Optimal order up-to level for the
supply chain participants
• Input
• Manufacturer have customer demand data sets of previous periods.
• Supply chain comprising five stage (Customer- Retailer – Warehouse-
Distributor -Manufacturer )
• There are two types of transportation facilities for all members of the SC,
thus two optimal levels are assigned for each members.
• Per unit back order/ Penalty cost, Per unit inventory holding cost, Per unit
sale price of the product, Per unit transportation costs for slow and fast
transportation, replenishment lead time of the transportations
• Supply chain modeling using optimal order up-to level
• Mathematical formulation (Objective function)
• Encoding (String representation)
• Manipulation of fitness value of the string
• Utilization of evolutionary algorithm to search best string
• Output (Optimal order up-to level which can provide most profitable
replenishment program)
 Supply chain model using optimal
order up-to level approach

Lead time Lead time Lead time

Customer Order
Order Order
demand
Ware-
Retailer Distributor Manufacture
house
Order Order Order
replenishment replenishment replenishment

Receive the
product that is
• Ordered quantity at the
in
starting of period =
transportation
(Optimal order up-to level)
– (Inventory at the end of
previous period)
Objective function
Total profit acquired by entire SC of M players, after running it for T period/weeks is

t T 

k M 
 ( SP  Fcdt )    HC× Linvk,t + BC× Binvk,t + TC1 ×Y1,k,t  TC2  Y2,k,t 

t 1  k 1 


Acquired Back order

Holding cast cast Transportation
revenue
cast

 Where
Linvk,t = Inventory at the end of the period t at the stage k
Binvk,t= Back order inventory that is not fulfilled
Y 1,k,t= Order created by the participant k using fast transportation
Y 2,k,t= Order created by the participant k using slow transportation
Fcdt= Fulfilled customer demand
HC, BC, TC1, TC2 per unit cost of holding, back order and
transportations respectively, SP = Per unit sale price of the product
Encoding

• Encoding is the first and foremost step in solving a problem

• To encipher a representation scheme for fixed order up-to inventory
levels, an Chromosome (candidate solution) is represented in a form of
string.
10 10 14 43 31 41

Chromosome  Candidate Solution

1. Numbers (1-6) represent the order up-to inventory levels i.e.
2. First three cells represents the first order up-to level for retailer,
warehouse and distributor respectively.
3. 4th , 5th and 6th represents the second order up-to level for retailer ,
warehouse and distributor respectively.
Supply chain using fixed order up-to
level

Customer
Demand Retailer Warehouse Distributor Manufacturer

•Order
•Order at the starting at the
of the starting
period •Order of at
thethe
period
starting
Order using fast It isofassumed
transportation
the period
= (Firstthat manufacturer
order up-to is able to satisfy all
level)-
Order using fast transportation = (First Orderorderusing
the of up-to
fast level)-
demand transportation
generated by=the
(First order up-to level)-
Distributor
(Inventory level
(Inventory level at the end of the period at the end
(Inventory the period t-1)
t-1) level at the end of the period t-1)
Order using slow
Order using slow transportation= (Second transportation=
Order using orderslowup-to(Second order up-to
transportation= (Second order up-to
level –at(the
level) – ( Inventorlevel) Inventor oflevel
end level) at( Inventor
the t-1
the –period end +oflevel
the period t-1 of
at the end + the period t-1 +
quantity replenished quantity
using replenished usingreplenished
first transportation)
quantity first transportation)
using first transportation)
•Inventory level at
•Inventory level at the starting of period the starting
•Inventory= Inventoryof period
level at the
level = Inventory
starting
at the end
of periodlevel=atInventory
the end level at the end
of + transported
of + transported inventory using fastinventory using fast
of transportation
+ transported transportation
inventory
in period (t-1)+
using fast in transportation
period (t-1)+ in period (t-1)+
transported
transported inventory using slowinventory using slow
transportation
transported transportation
inventory
in period using
(t-2) slow in transportation
period (t-2) in period (t-2)
•Satisfy demand
•Fulfillment of customer the orderin created
•Satisfy by order
the period
the the
t retailer
created in by
thewarehouse
period t in the period t
 How supply chain are simulated
using fixed order up-to level
Let, released inventory from warehouse using slow transportation at period of (t-2)=26,
Released inventory from warehouse in the starting of the period (t-1)
using fast transportation =0,
using slow transportation =29,
Inventory level at the end of period (t-1)=8
If order up to levels represented as string 10 10 14 43 31 41

Replenishment processing at retailer stage in period t:

Custom Period Initial Fulfille Remaine Back Deman Reple Deman Replen Initial
er Invento d d order d using nishe d using ished inventor
Deman ry deman Inventor fast d slow quantit y
d (Retail d y at the transpo quanti transpor y (Wareh
er) end ty t ouse)
rt
= 8+26+0
33 tth 34 33 = 34-331 0 2 2 33 27 29
= 10 - 8 = 43-(8+2)
Customer demand <
Inventory available Demands are fully
= 29>2 = 27 out of 33 has been
satisfied
fulfilled
 Continue…
Custom Period Initial Fulfille Remai Back Deman Reple Demand Repleni Initial
er Invento d ned order d using nished order shed invento
Deman ry deman Invento fast quantit using quantity ry
d (Retaile d ry at transpor y slow (Wareh
r) the end transpor ouse)
t tation

33 tth 34 33 1 0 2 2 33 27 29
26 (t+1)t 32 26 6 0 9 9 33 20 29
h

30 (t+2)t 42 30 12 0 4 4 33 27 31
h
This column contribute to provide
Incurred costs (back order cost, inventory holding
… … …
sum of acquired revenue
… … … … … … …
cost, transportation cost) at retailer stage
…
Example

 In order to assign the optimal order up-to level (s) manufacturer

have information
 Period wise customer demand pattern =Normally distributed with
mean=30, variance= 5.0
 Per unit back order cost =1.5 $
 Per unit sale price of product=3.0 $
 Per unit inventory holding cost= 0.25 $
 Per unit transportation cost (fast transportation, lead time one
period )= 0.4 $
 Per unit transportation cost (slow transportation, lead time two
period )= 0.2 $
Manipulation of optimal order up-to
levels
 Generate the several set of customer demand data on the
basis of demand distribution.
 For all sets of customer demand optimal order up-to levels
are determined using random optimization technique.
 The result (order up-to level) which is suitable to provide
maximum profit in most of experiments that are selected as
a optimal order up-to levels.
 Acquire optimal order up-to levels have been tested over
practical data set of customer demand, performance of
supply chain are observed in form of customer satisfaction,
and period wise inventory level at the each stages.
 Experiments conducted
Experiment Maximum Profit Optimum order up-to levels

1 4361.5 <14, 10, 17, 41, 30, 36 >

2 4515.92 <10,10, 11, 43, 31, 36 >

3 5015.1 <10,10, 11, 43, 31, 36 >

4 4550.6 <10, 10, 13, 43, 31, 37 >

5 4508.96 <12,10,14, 41, 30, 37 >

6 4601.9 <12,10,14, 41, 30, 37 >

For all these six
7 4455.1 <10,10,11, 33, 29, 35 > experiments results (10, 10,
8 4491.92 <10,10,11, 33, 29, 35 > 14, 43, 31, 41)
9 4305.96 <10,10,11, 33, 29, 35 > is a same, thus has been
10 4732.351 <10, 10, 14, 43, 31,41> selected as a suitable
11 4517.73 <10, 10, 14, 43, 31, 41> optimal up-to level for
12 4747.966 <10, 10, 14, 43, 31, 41> given SC scenario
13 4795.146 <10, 10, 14, 43, 31, 41>

14 4591.314 <10, 10, 14, 43, 31, 41>

15 5190.5 <10, 10, 14, 43, 31, 41>

16 4539.8 <14, 10, 17, 41, 30, 36 >

17 4471.8 <13, 10, 11, 40, 31, 37 >

18 4553.7 <13, 10, 11, 38, 31, 37 >

19 4470.1 <11, 10, 11, 41, 31, 39 >

20 4603.0 <11, 13, 12, 41, 31,39 >
 Test over practical data set
Performance of supply chain in form Inventory level at each stages by employing
optimal order up-to level (10, 10, 14, 43, 31, 41)

Inventory level (R etailer)

Inventory level (Warehous e)
Inventory level (D is tributor)
C us tomer D emand
60
50
Inventory levell

40
30
20
10
0
1 7 13 19 25 31 37 43 49 55 61 67 73 79 85 91 97
Number of period
Performance in form of customer
satisfaction

P erc entage of fulfilled

c us tomer demand

120
100
P ercentage

80
60
40
20
0
1 7 13 19 25 31 37 43 49 55 61 67 73 79 85 91 97

Number of period

Results show that in most of the periods

customer demands are fully satisfied
Importance of Optimal order up-to level
bashed approach
 It provides a centralized approach by considering
supply chain as a whole.
 Optimal order up-to level (s) based monitoring of
product replenishments facilitates a stringent
control over supply line inventory level and
customer satisfaction.
 It facilitates the economized utilization of
transportation options to provide a steady flow of
material through out the supply chain.
Summary of Learning Objectives
 What are supply chain coordination and the
bullwhip effect, and what are their effects on
supply chain performance?
 What are the causes of the bullwhip effect,
and what are obstacles to coordination in the
supply chain?
 What are the managerial levers that help
achieve coordination in the supply chain?
 What are actions that facilitate the building of
strategic partnerships and trust in the supply
chain?