Chapter 6

Inventory Control Models

To accompany
Quantitative Analysis for Management, Tenth Edition,
by Render, Stair, and Hanna © 2008 Prentice-Hall, Inc.
Power Point slides created by Jeff Heyl © 2009 Prentice-Hall, Inc.
 Learning Objectives
After completing this chapter, students will be able to:

1. Understand the importance of inventory

control and ABC analysis
2. Use the economic order quantity (EOQ) to
determine how much to order
3. Compute the reorder point (ROP) in
determining when to order more inventory
4. Handle inventory problems that allow
quantity discounts or noninstantaneous
receipt

© 2009 Prentice-Hall, Inc. 6–2

 Learning Objectives
After completing this chapter, students will be able to:

5. Understand the use of safety stock with

known and unknown stockout costs
6. Describe the use of material requirements
planning in solving dependent-demand
inventory problems
7. Discuss just-in-time inventory concepts to
reduce inventory levels and costs
8. Discuss enterprise resource planning
systems

© 2009 Prentice-Hall, Inc. 6–3

 Chapter Outline
6.1 Introduction
6.2 Importance of Inventory Control
6.3 Inventory Decisions
6.4 Economic Order Quantity: Determining
How Much to Order
6.5 Reorder Point: Determining When to Order
6.6 EOQ Without the Instantaneous Receipt
Assumption
6.7 Quantity Discount Models
6.8 Use of Safety Stock

© 2009 Prentice-Hall, Inc. 6–4

 Chapter Outline
6.9 Single-Period Inventory Models
6.10 ABC Analysis
6.11 Dependent Demand: The Case for Material
Requirements Planning
6.12 Just-in-Time Inventory Control
6.13 Enterprise Resource Planning

© 2009 Prentice-Hall, Inc. 6–5

 Introduction
 Inventory is an expensive and important
asset to many companies
 Lower inventory levels can reduce costs
 Low inventory levels may result in stockouts
and dissatisfied customers
 Most companies try to balance high and low
inventory levels with cost minimization as a
goal
 Inventory is any stored resource used to
satisfy a current or future need
 Common examples are raw materials, work-
in-process, and finished goods

© 2009 Prentice-Hall, Inc. 6–6

 Introduction
 Inventory may account for 50% of the total
invested capital of an organization and 70% of the
cost of goods sold
Capital
Energy Costs
Costs

Labor Costs

Inventory
Costs

© 2009 Prentice-Hall, Inc. 6–7

 Introduction
 All organizations have some type of inventory
control system
 Inventory planning helps determine what
goods and/or services need to be produced
 Inventory planning helps determine whether
the organization produces the goods or
services or whether they are purchased from
another organization
 Inventory planning also involves demand
forecasting

© 2009 Prentice-Hall, Inc. 6–8

 Introduction
 Inventory planning and control

Planning on What
Forecasting Controlling
Inventory to Stock
Parts/Product Inventory
and How to Acquire
Demand Levels
It

Feedback Measurements
to Revise Plans and
Forecasts

Figure 6.1

© 2009 Prentice-Hall, Inc. 6–9

 Importance of Inventory Control

 Five uses of inventory

 The decoupling function
 Storing resources
 Irregular supply and demand
 Quantity discounts
 Avoiding stockouts and shortages
 The decoupling function
 Used as a buffer between stages in a
manufacturing process
 Reduces delays and improves efficiency

© 2009 Prentice-Hall, Inc. 6 – 10

 Importance of Inventory Control

 Storing resources
 Seasonal products may be stored to satisfy
off-season demand
 Materials can be stored as raw materials,
work-in-process, or finished goods
 Labor can be stored as a component of
partially completed subassemblies
 Irregular supply and demand
 Demand and supply may not be constant
over time
 Inventory can be used to buffer the variability

© 2009 Prentice-Hall, Inc. 6 – 11

 Importance of Inventory Control

 Quantity discounts
 Lower prices may be available for larger
orders
 Extra costs associated with holding more
inventory must be balanced against lower
purchase price
 Avoiding stockouts and shortages
 Stockouts may result in lost sales
 Dissatisfied customers may choose to buy
from another supplier

© 2009 Prentice-Hall, Inc. 6 – 12

 Inventory Decisions
 There are only two fundamental decisions
in controlling inventory
 How much to order
 When to order
 The major objective is to minimize total
inventory costs
 Common inventory costs are
 Cost of the items (purchase or material cost)
 Cost of ordering
 Cost of carrying, or holding, inventory
 Cost of stockouts

© 2009 Prentice-Hall, Inc. 6 – 13

 Inventory Cost Factors
ORDERING COST FACTORS CARRYING COST FACTORS
Developing and sending purchase orders Cost of capital
Processing and inspecting incoming
Taxes
inventory
Bill paying Insurance

Inventory inquiries Spoilage

Utilities, phone bills, and so on, for the
Theft
purchasing department
Salaries and wages for the purchasing
Obsolescence
department employees
Supplies such as forms and paper for the Salaries and wages for warehouse
purchasing department employees
Utilities and building costs for the
warehouse
Supplies such as forms and paper for the
warehouse

Table 6.1
© 2009 Prentice-Hall, Inc. 6 – 14
 Inventory Cost Factors
 Ordering costs are generally independent
of order quantity
 Many involve personnel time
 The amount of work is the same no matter the
size of the order
 Carrying costs generally varies with the
amount of inventory, or the order size
 The labor, space, and other costs increase as
the order size increases
 Of course, the actual cost of items
purchased varies with the quantity
purchased
© 2009 Prentice-Hall, Inc. 6 – 15
 Economic Order Quantity

 The economic order quantity (EOQ) model

is one of the oldest and most commonly
known inventory control techniques
 It dates from 1915
 It is easy to use but has a number of
important assumptions

© 2009 Prentice-Hall, Inc. 6 – 16

 Economic Order Quantity
 Assumptions
1. Demand is known and constant
2. Lead time is known and constant
3. Receipt of inventory is instantaneous
4. Purchase cost per unit is constant
throughout the year
5. The only variable costs are the placing an
order, ordering cost, and holding or storing
inventory over time, holding or carrying cost,
and these are constant throughout the year
6. Orders are placed so that stockouts or
shortages are avoided completely

© 2009 Prentice-Hall, Inc. 6 – 17

 Inventory Usage Over Time

Inventory
Level
Order Quantity = Q =
Maximum Inventory Level

Minimum
Inventory

0
Time

Figure 6.2
© 2009 Prentice-Hall, Inc. 6 – 18
 Inventory Costs in the EOQ Situation

 Objective is generally to minimize total cost

 Relevant costs are ordering costs and carrying
costs
Q
Average inventory level 
2
INVENTORY LEVEL
DAY BEGINNING ENDING AVERAGE
April 1 (order received) 10 8 9
April 2 8 6 7
April 3 6 4 5
April 4 4 2 3
April 5 2 0 1
Maximum level April 1 = 10 units
Total of daily averages = 9 + 7 + 5 + 3 + 1 = 25
Number of days = 5
Average inventory level = 25/5 = 5 units Table 6.2

© 2009 Prentice-Hall, Inc. 6 – 19

 Inventory Costs in the EOQ Situation

 Mathematical equations can be developed using

Q = number of pieces to order

EOQ = Q* = optimal number of pieces to order
D = annual demand in units for the inventory item
Co = ordering cost of each order
Ch = holding or carrying cost per unit per year

Number of Ordering
Annual ordering cost  orders placed  cost per
per year order
D
 Co
Q

© 2009 Prentice-Hall, Inc. 6 – 20

 Inventory Costs in the EOQ Situation

 Mathematical equations can be developed using

Q = number of pieces to order

EOQ = Q* = optimal number of pieces to order
D = annual demand in units for the inventory item
Co = ordering cost of each order
Ch = holding or carrying cost per unit per year

Average Carrying
Annual holding cost  inventory  cost per unit
per year

Q
 Ch
2

© 2009 Prentice-Hall, Inc. 6 – 21

 Inventory Costs in the EOQ Situation

Cost
Curve of Total Cost
of Carrying
and Ordering

Minimum
Total
Cost

Carrying Cost Curve

Ordering Cost Curve

Optimal Order Quantity

Order
Figure 6.3 Quantity

© 2009 Prentice-Hall, Inc. 6 – 22

 Finding the EOQ
 When the EOQ assumptions are met, total cost is
minimized when Annual ordering cost = Annual
holding cost
D Q
Co  Ch
Q 2
 Solving for Q
2 DC o  Q 2C h
2 DC o
 Q2
Ch
2 DC o
 Q  EOQ  Q *
Ch
© 2009 Prentice-Hall, Inc. 6 – 23
Economic Order Quantity (EOQ) Model

 Summary of equations

D
Annual ordering cost  C o
Q

Q
Annual holding cost  C h
2

2 DC o
EOQ  Q * 
Ch

© 2009 Prentice-Hall, Inc. 6 – 24

 Sumco Pump Company Example

 Company sells pump housings to other

companies
 Would like to reduce inventory costs by finding
optimal order quantity
 Annual demand = 1,000 units
 Ordering cost = $10 per order
 Average carrying cost per unit per year = $0.50

2 DC o 2(1,000)(10)
Q 
*
  40,000  200 units
Ch 0.50

© 2009 Prentice-Hall, Inc. 6 – 25

Sumco Pump Company Example

Total annual cost = Order cost + Holding cost

D Q
TC  C o  C h
Q 2
1,000 200
 (10)  (0.5 )
200 2
 $50  $50  $100

© 2009 Prentice-Hall, Inc. 6 – 26

Sumco Pump Company Example

Program 6.1A
© 2009 Prentice-Hall, Inc. 6 – 27
Sumco Pump Company Example

Program 6.1B
© 2009 Prentice-Hall, Inc. 6 – 28
 Purchase Cost of Inventory Items
 Total inventory cost can be written to include the
cost of purchased items
 Given the EOQ assumptions, the annual purchase
cost is constant at D  C no matter the order
policy
 C is the purchase cost per unit
 D is the annual demand in units
 It may be useful to know the average dollar level
of inventory

(CQ )
Average dollar level 
2
© 2009 Prentice-Hall, Inc. 6 – 29
 Purchase Cost of Inventory Items
 Inventory carrying cost is often expressed as an
annual percentage of the unit cost or price of the
inventory
 This requires a new variable

Annual inventory holding charge as

I a percentage of unit price or cost
 The cost of storing inventory for one year is then

C h  IC

2 DC o
thus, Q 
*

IC
© 2009 Prentice-Hall, Inc. 6 – 30
 Sensitivity Analysis with the
EOQ Model
 The EOQ model assumes all values are know and
fixed over time
 Generally, however, the values are estimated or
may change
 Determining the effects of these changes is
called sensitivity analysis
 Because of the square root in the formula,
changes in the inputs result in relatively small
changes in the order quantity

2 DC o
EOQ 
Ch

© 2009 Prentice-Hall, Inc. 6 – 31

 Sensitivity Analysis with the
EOQ Model
 In the Sumco example

2(1,000)(10)
EOQ   200 units
0.50

 If the ordering cost were increased four times from

$10 to $40, the order quantity would only double

2(1,000)(40)
EOQ   400 units
0.50

 In general, the EOQ changes by the square root

of a change to any of the inputs

© 2009 Prentice-Hall, Inc. 6 – 32

 Reorder Point:
Determining When To Order
 Once the order quantity is determined, the
next decision is when to order
 The time between placing an order and its
receipt is called the lead time (L) or
delivery time
 When to order is generally expressed as a
reorder point (ROP)

Demand Lead time for a

ROP  per day  new order in days

dL
© 2009 Prentice-Hall, Inc. 6 – 33
 Procomp’s Computer Chip Example

 Demand for the computer chip is 8,000 per year

 Daily demand is 40 units
 Delivery takes three working days

ROP  d  L  40 units per day  3 days

 120 units

 An order is placed when the inventory reaches

120 units
 The order arrives 3 days later just as the
inventory is depleted

© 2009 Prentice-Hall, Inc. 6 – 34

 EOQ Without The Instantaneous
Receipt Assumption
 When inventory accumulates over time, the
instantaneous receipt assumption does not apply
 Daily demand rate must be taken into account
 The revised model is often called the production
run model

Inventory
Level Part of Inventory Cycle There is No Production
During Which Production is During This Part of the
Taking Place Inventory Cycle
Maximum
Inventory

t Time

Figure 6.5
© 2009 Prentice-Hall, Inc. 6 – 35
 Annual Carrying Cost for
Production Run Model
 In production runs, setup cost replaces ordering
cost
 The model uses the following variables

Q  number of pieces per order, or

production run
Cs  setup cost
Ch  holding or carrying cost per unit per
year
p  daily production rate
d  daily demand rate
t  length of production run in days

© 2009 Prentice-Hall, Inc. 6 – 36

 Annual Carrying Cost for
Production Run Model
Maximum inventory level
 (Total produced during the production run)
– (Total used during the production run)
 (Daily production rate)(Number of days production)
– (Daily demand)(Number of days production)
 (pt) – (dt)

since Total produced  Q  pt

Q
we know t
p
Maximum Q Q  d
inventory  pt  dt  p  d  Q 1  
level p p  p
© 2009 Prentice-Hall, Inc. 6 – 37
 Annual Carrying Cost for
Production Run Model
 Since the average inventory is one-half the
maximum

Q d 
Average inventory   1 
2 p

and

Q d 
Annual holding cost   1  C h
2 p

© 2009 Prentice-Hall, Inc. 6 – 38

 Annual Setup Cost for
Production Run Model
 Setup cost replaces ordering cost when a product
is produced over time

D
Annual setup cost  C s
Q

and
D
Annual ordering cost  C o
Q

© 2009 Prentice-Hall, Inc. 6 – 39

 Determining the Optimal
Production Quantity
 By setting setup costs equal to holding costs, we
can solve for the optimal order quantity

Annual holding cost  Annual setup cost

Q d  D
 1   h
C  Cs
2 p Q

 Solving for Q, we get

2 DC s
Q* 
 d
C h  1 
 p
© 2009 Prentice-Hall, Inc. 6 – 40
 Production Run Model

 Summary of equations

Q d 
Annual holding cost   1  C h
2 p

D
Annual setup cost  C s
Q

2 DC s
Optimal production quantity Q * 
 d
C h  1 
 p

© 2009 Prentice-Hall, Inc. 6 – 41

 Brown Manufacturing Example

 Brown Manufacturing produces commercial

refrigeration units in batches

Annual demand  D  10,000 units

Setup cost  Cs  $100
Carrying cost  Ch  $0.50 per unit per year
Daily production rate  p  80 units daily
Daily demand rate  d  60 units daily

© 2009 Prentice-Hall, Inc. 6 – 42

 Brown Manufacturing Example

2 DC s Q
1. Q 
*
Production cycle 
 d p
C h  1 
 p 4,000
  50 days
80
2  10,000  100
2. Q 
*

 60 
0.5 1  
 80 
2,000,000
  16,000,000
 
0.5 1
4
 4,000 units
© 2009 Prentice-Hall, Inc. 6 – 43
 Brown Manufacturing Example

Program 6.2A
© 2009 Prentice-Hall, Inc. 6 – 44
 Brown Manufacturing Example

Program 6.2B
© 2009 Prentice-Hall, Inc. 6 – 45
 Quantity Discount Models
 Quantity discounts are commonly available
 The basic EOQ model is adjusted by adding in the
purchase or materials cost

Total cost  Material cost + Ordering cost + Holding cost

D Q
Total cost  DC  C o  C h
Q 2
where
D  annual demand in units
Cs  ordering cost of each order
C  cost per unit
Ch  holding or carrying cost per unit per year
© 2009 Prentice-Hall, Inc. 6 – 46
 Quantity Discount Models
Because
 Quantity unitare
discounts cost is now variable
commonly available
Holding
 The basic EOQ modelcost  Ch  IC
is adjusted by adding in the
purchase
I  holdingorcost
materials cost
as a percentage of the unit cost (C)
Total cost  Material cost + Ordering cost + Holding cost
D Q
Total cost  DC  C o  C h
Q 2
where
D  annual demand in units
Cs  ordering cost of each order
C  cost per unit
Ch  holding or carrying cost per unit per year
© 2009 Prentice-Hall, Inc. 6 – 47
 Quantity Discount Models
 A typical quantity discount schedule

DISCOUNT DISCOUNT DISCOUNT

NUMBER QUANTITY DISCOUNT (%) COST ($)
1 0 to 999 0 5.00

2 1,000 to 1,999 4 4.80

3 2,000 and over 5 4.75

Table 6.3

 Buying at the lowest unit cost is not always the

best choice

© 2009 Prentice-Hall, Inc. 6 – 48

 Quantity Discount Models
 Total cost curve for the quantity discount model

Total TC Curve for Discount 3

Cost
$ TC Curve for
Discount 1

TC Curve for Discount 2

EOQ for Discount 2

0 1,000 2,000
Figure 6.6 Order Quantity
© 2009 Prentice-Hall, Inc. 6 – 49
 Brass Department Store Example
 Brass Department Store stocks toy race cars
 Their supplier has given them the quantity
discount schedule shown in Table 6.3
 Annual demand is 5,000 cars, ordering cost is $49, and
holding cost is 20% of the cost of the car
 The first step is to compute EOQ values for each
discount
(2)(5,000)(49)
EOQ1   700 cars per order
(0.2)(5.00)
(2)(5,000)(49)
EOQ 2   714 cars per order
(0.2)(4.80)
(2)(5,000)(49)
EOQ 3   718 cars per order
(0.2)(4.75 )
© 2009 Prentice-Hall, Inc. 6 – 50
 Brass Department Store Example

 The second step is adjust quantities below the

allowable discount range
 The EOQ for discount 1 is allowable
 The EOQs for discounts 2 and 3 are outside the
allowable range and have to be adjusted to the
smallest quantity possible to purchase and
receive the discount

Q1  700
Q2  1,000
Q3  2,000

© 2009 Prentice-Hall, Inc. 6 – 51

 Brass Department Store Example

 The third step is to compute the total cost for

each quantity
ANNUAL ANNUAL ANNUAL
UNIT ORDER MATERIAL ORDERING CARRYING
DISCOUNT PRICE QUANTITY COST ($) COST ($) COST ($)
NUMBER (C) (Q) = DC = (D/Q)Co = (Q/2)Ch TOTAL ($)

1 $5.00 700 25,000 350.00 350.00 25,700.00

2 4.80 1,000 24,000 245.00 480.00 24,725.00

3 4.75 2,000 23,750 122.50 950.00 24,822.50

Table 6.4

 The fourth step is to choose the alternative

with the lowest total cost
© 2009 Prentice-Hall, Inc. 6 – 52
Brass Department Store Example

Program 6.3A
© 2009 Prentice-Hall, Inc. 6 – 53
Brass Department Store Example

Program 6.3B

© 2009 Prentice-Hall, Inc. 6 – 54

 Use of Safety Stock

 If demand or the lead time are uncertain,

the exact ROP will not be known with
certainty
 To prevent stockouts, it is necessary to
carry extra inventory called safety stock
 Safety stock can prevent stockouts when
demand is unusually high
 Safety stock can be implemented by
adjusting the ROP

© 2009 Prentice-Hall, Inc. 6 – 55

 Use of Safety Stock
 The basic ROP equation is
ROP  d  L
d  daily demand (or average daily demand)
L  order lead time or the number of
working days it takes to deliver an order
(or average lead time)

 A safety stock variable is added to the equation

to accommodate uncertain demand during lead
time
ROP  d  L + SS
where
SS  safety stock

© 2009 Prentice-Hall, Inc. 6 – 56

 Use of Safety Stock
Inventory
on Hand

Time
Figure 6.7(a) Stockout
© 2009 Prentice-Hall, Inc. 6 – 57
 Use of Safety Stock
Inventory
on Hand

Safety
Stock, SS
Stockout is Avoided

Time
Figure 6.7(b)
© 2009 Prentice-Hall, Inc. 6 – 58
 ROP with Known Stockout Costs

 With a fixed EOQ and a ROP for placing orders,

stockouts can only occur during lead time
 Objective is to find the safety stock quantity that
will minimize the total of stockout cost and
holding cost
 Need to know the stockout cost per unit and the
probability distribution of demand during lead
time
 Estimating stockout costs can be difficult as
there are direct and indirect costs

© 2009 Prentice-Hall, Inc. 6 – 59

 ABCO Example

 ABCO, Inc. has determined its ROP is 50 units

 The carrying cost per unit is $40
 The probability distribution of demand during
lead time is shown below
NUMBER OF UNITS PROBABILITY
30 0.2
40 0.2
ROP 50 0.3
60 0.2
70 0.1
1.0
Table 6.5
© 2009 Prentice-Hall, Inc. 6 – 60
 ABCO Example
 Determining the safety stock level that will
minimize total expected cost is a decision making
under risk problem
PROBABILITY 0.20 0.20 0.30 0.20 0.10
STATE OF
NATURE DEMAND DURING LEAD TIME
ALTERNATIVE 30 40 50 60 70 EMV
ROP 30 $ 0 $2,400 $4,800 $7,200 $9,600 $4,320
40 50 0 2,400 4,800 7,200 2,410
50 100 50 0 2,400 4,800 990
60 150 100 50 0 2,400 305
70 200 150 100 50 0 110

Table 6.6

© 2009 Prentice-Hall, Inc. 6 – 61

 ABCO Example

 When the ROP is less than demand over lead

time, total cost is equal to stockout cost

Total cost  Stockout cost  Number of units short

 Stockout cost per unit  Number of orders per year

 When the ROP is greater than demand over lead

time, total costs will be equal to total additional
carrying costs

Total cost  Total additional carrying cost

 Number of surplus units  Carrying cost

© 2009 Prentice-Hall, Inc. 6 – 62

 ABCO Example

 EMVs for each reorder point

$5,000 –
$4,320
$4,000 –

$3,000 –
Cost

$2,410
$2,000 –

$990
$1,000 –
$305 $110
$0 –
30 Units 40 Units 50 Units 60 Units 70 Units
ROP
Figure 6.8
© 2009 Prentice-Hall, Inc. 6 – 63
 Safety Stock with Unknown
Stockout Costs
 There are many situations when stockout costs
are unknown
 An alternative approach to determining safety
stock levels is to use a service level
 A service level is the percent of time you will not
be out of stock of a particular item

Service level  1 – Probability of a stockout

or
Probability of a stockout  1 – Service level

© 2009 Prentice-Hall, Inc. 6 – 64

 Hinsdale Company Example
 Inventory demand is normally distributed during
the reorder period
 Mean is 350 units and standard deviation is 10
 They want stockouts to occur only 5% of the time

  Mean demand  350

  Standard deviation  10
X  Mean demand + Safety stock
SS  Safety stock  X –   Z
X  5% Area of
Z Normal Curve
 SS

  350 X?
Figure 6.9

© 2009 Prentice-Hall, Inc. 6 – 65

 Hinsdale Company Example
X  SS
 From Appendix A we find Z  1.65  
 
 Solving for safety stock

SS  1.65(10)  16.5 units, or 17 units

  Mean demand  350

  Standard deviation  10
X  Mean demand + Safety stock
SS  Safety stock  X –   Z
X  5% Area of
Z Normal Curve
 SS

  350 X?
Figure 6.9

© 2009 Prentice-Hall, Inc. 6 – 66

 Hinsdale Company Example
 Different safety stock levels will be generated for
different service levels
 However, the relationship is not linear
 You should be aware of what a service level is costing in
terms of carrying the safety stock in inventory
 The relationship between Z and safety stock can
be developed as follows

X 
1. We know that Z 
 4. So we have
2. We also know that SS  X –  SS  Z
 Z(10)
SS
3. Thus Z 

© 2009 Prentice-Hall, Inc. 6 – 67
 Hinsdale Company Example
 Cost of different service levels
SERVICE Z VALUE FROM SAFETY STOCK CARRYING
LEVEL (%) NORMAL CURVE TABLE (UNITS) COST ($)
90 1.28 12.8 12.80
91 1.34 13.4 13.40
92 1.41 14.1 14.10
93 1.48 14.8 14.80
94 1.55 15.5 15.50
95 1.65 16.5 16.50
96 1.75 17.5 17.50
97 1.88 18.8 18.80
98 2.05 20.5 20.50
99 2.33 23.3 23.20
99.99 3.72 37.2 37.20

Table 6.7
© 2009 Prentice-Hall, Inc. 6 – 68
 Hinsdale Company Example
 Service level ($)
40 –
–
versus annual –
–
–
carrying cost 35 –
–
–

Inventory Carrying Costs ($)

–
 This was –
30 –
–
developed for –
–
–
a specific 25 –
–
–
case, but the –
–
20 –
general shape –
–
–
of the curve is –
15 –
–
the same for –
–
all service- –
10 –
–
level problems –
–
– | | | | | | | | | | |
–
90 91 92 93 94 95 96 97 98 99 99.99 (%)
Figure 6.8 Service Level (%)

© 2009 Prentice-Hall, Inc. 6 – 69

 Single-Period Inventory Models

 Some products have no future value beyond the

current period
 These situations are called news vendor
problems or single-period inventory models
 Analysis uses marginal profit (MP) and marginal
loss (ML) and is called marginal analysis
 With a manageable number of states of nature
and alternatives, discrete distributions can be
used
 When there are a large number of alternatives or
states of nature, the normal distribution may be
used

© 2009 Prentice-Hall, Inc. 6 – 70

 Marginal Analysis with
Discrete Distributions
 We stock an additional unit only if the expected
marginal profit for that unit exceeds the expected
marginal loss
P  probability that demand will be greater
than or equal to a given supply (or the
probability of selling at least one
additional unit)

1 – P  probability that demand will be less than

supply (or the probability that one
additional unit will not sell)

© 2009 Prentice-Hall, Inc. 6 – 71

 Marginal Analysis with
Discrete Distributions
 The expected marginal profit is P(MP)
 The expected marginal loss is (1 – P)(ML)
 The optimal decision rule is to stock the
additional unit if
P(MP) ≥ (1 – P)ML
 With some basic manipulation

P(MP) ≥ ML – P(ML) or
ML
P(MP) + P(ML) ≥ ML P
ML  MP
P(MP + ML) ≥ ML

© 2009 Prentice-Hall, Inc. 6 – 72

 Steps of Marginal Analysis with
Discrete Distributions
ML
1. Determine the value of for the
problem ML  MP
2. Construct a probability table and add a
cumulative probability column
3. Keep ordering inventory as long as the
probability (P) of selling at least one additional
ML
unit is greater than
ML  MP

© 2009 Prentice-Hall, Inc. 6 – 73

 Café du Donut Example
 The café buys donuts each day for $4 per carton
of 2 dozen donuts
 Any cartons not sold are thrown away at the end
of the day
 If a carton is sold, the total revenue is $6
 The marginal profit per carton is

MP  Marginal profit  $6 – $4  $2

 The marginal loss is $4 per carton since cartons

can not be returned or salvaged

© 2009 Prentice-Hall, Inc. 6 – 74

 Café du Donut Example
 Probability distribution of daily demand

DAILY SALES PROBABILITY (P) THAT DEMAND

(CARTONS OF DOUGHNUTS) WILL BE AT THIS LEVEL
4 0.05
5 0.15
6 0.15
7 0.20
8 0.25
9 0.10
10 0.10
Total 1.00

Table 6.8

© 2009 Prentice-Hall, Inc. 6 – 75

 Café du Donut Example
ML
Step 1. Determine the value of for the
decision rule ML  MP

ML $4 4
P    0.67
ML  MP $4  $2 6
P  0.67

Step 2. Add a new column to the table to reflect the

probability that doughnut sales will be at
each level or greater

© 2009 Prentice-Hall, Inc. 6 – 76

 Café du Donut Example
 Marginal analysis

DAILY SALES PROBABILITY (P) THAT PROBABILITY (P) THAT

(CARTONS OF DEMAND WILL BE AT DEMAND WILL BE AT THIS
DOUGHNUTS) THIS LEVEL LEVEL OR GREATER
4 0.05 1.00 ≥ 0.66
5 0.15 0.95 ≥ 0.66
6 0.15 0.80 ≥ 0.66
7 0.20 0.65
8 0.25 0.45
9 0.10 0.20
10 0.10 0.10
Total 1.00
Table 6.9

© 2009 Prentice-Hall, Inc. 6 – 77

 Café du Donut Example

Step 3. Keep ordering additional cartons as long

as the probability of selling at least one
additional carton is greater than P, which
is the indifference probability

P at 6 cartons  0.80 > 0.67

© 2009 Prentice-Hall, Inc. 6 – 78

 Marginal Analysis with the
Normal Distribution
 We first need to find four values

1. The average or mean sales for the product, 

2. The standard deviation of sales, 
3. The marginal profit for the product, MP
4. The marginal loss for the product, ML

 We let X *  optimal stocking level

© 2009 Prentice-Hall, Inc. 6 – 79

 Steps of Marginal Analysis with the
Normal Distribution
ML
1. Determine the value of for the
problem ML  MP
2. Locate P on the normal distribution (Appendix
A) and find the associated Z-value
3. Find X * using the relationship
X*  m
Z
s
to solve for the resulting stocking policy:
X *    Z

© 2009 Prentice-Hall, Inc. 6 – 80

 Newspaper Example
 Demand for the Chicago Tribune at Joe’s
Newsstand averages 60 papers a day with a
standard deviation of 10
 The marginal loss is 20 cents and the marginal
profit is 30 cents
Step 1. Joe should stock the Tribune as long as
the probability of selling the last unit is at least
ML/(ML + MP):
ML 20 cents 20
   0.40
ML  MP 20 cents  30 cents 50

Let P = 0.40
© 2009 Prentice-Hall, Inc. 6 – 81
 Newspaper Example
Step 2. Using the normal distribution in Figure
6.11, we find the appropriate Z value
Z = 0.25 standard deviations from the mean
Area under the Curve is 1 – 0.40 = 0.60
(Z = 0.25) Mean Daily Sales

Area under the Curve is 0.40

  50 X* X  Demand

Figure 6.11 Optimal Stocking Policy (62 Newspapers)

© 2009 Prentice-Hall, Inc. 6 – 82

 Newspaper Example
Step 3. In this problem,   60 and   10, so

X *  60
0.25 
10
or

X *  60 + 0.25(10)  62.5, or 62 newspapers

 Joe should order 62 newspapers since the

probability of selling 63 newspapers is
slightly less than 0.40

© 2009 Prentice-Hall, Inc. 6 – 83

 Newspaper Example
 The procedure is the same when P > 0.50
 Joe also stocks the Chicago Sun-Times
 Marginal loss is 40 cents and marginal profit is
10 cents
 Daily sales average 100 copies with a standard
deviation of 10 papers
ML 40 cents 40
   0.80
ML  MP 40 cents  10 cents 50

 From Appendix A

Z = – 0.84 standard deviations from the mean

© 2009 Prentice-Hall, Inc. 6 – 84

 Newspaper Example
With   100 and   10
X *  100
 0.84 
10
Area under the
or Curve is 0.40
(Z = – 0.84)
X *  100 – 0.84(10)
 91.6, or 91 newspapers

Figure 6.11 X *   100 X  Demand

Optimal Stocking Policy (91 Newspapers)

© 2009 Prentice-Hall, Inc. 6 – 85
 ABC Analysis
 The purpose of ABC analysis is to divide the
inventory into three groups based on the overall
inventory value of the items
 Group A items account for the major portion of
inventory costs
 Typically about 70% of the dollar value but only 10% of
the quantity of items
 Forecasting and inventory management must be done
carefully
 Group B items are more moderately priced
 May represent 20% of the cost and 20% of the quantity
 Group C items are very low cost but high volume
 It is not cost effective to spend a lot of time managing
these items

© 2009 Prentice-Hall, Inc. 6 – 86

 ABC Analysis

 Summary of ABC analysis

INVENTORY DOLLAR INVENTORY ARE QUANTITATIVE CONTROL

GROUP USAGE (%) ITEMS (%) TECHNIQUES USED?

A 70 10 Yes

B 20 20 In some cases

C 10 70 No

Table 6.10

© 2009 Prentice-Hall, Inc. 6 – 87

 Dependent Demand: The Case for
Material Requirements Planning

 All the inventory models discussed so far

have assumed demand for one item is
independent of the demand for any other
item
 However, in many situations items
demand is dependent on demand for one
or more other items
 In these situations, MRP can be employed
effectively

© 2009 Prentice-Hall, Inc. 6 – 88

 Dependent Demand: The Case for
Material Requirements Planning

 Some of the benefits of MRP are

1. Increased customer service levels
2. Reduced inventory costs
3. Better inventory planning and scheduling
4. Higher total sales
5. Faster response to market changes and shifts
6. Reduced inventory levels without reduced
customer service
 Most MRP systems are computerized, but
the basic analysis is straightforward

© 2009 Prentice-Hall, Inc. 6 – 89

 Material Structure Tree
 The first step is to develop a bill of materials
(BOM)
 The BOM identifies components, descriptions,
and the number required for production of one
unit of the final product
 From the BOM we can develop a material
structure tree
 We use the following data
 Demand for product A is 50 units
 Each A requires 2 units of B and 3 units of C
 Each B requires 2 units of D and 3 units of E
 Each C requires 1 unit of E and 2 units of F

© 2009 Prentice-Hall, Inc. 6 – 90

 Material Structure Tree
Material structure tree for Item A
Level
0 A

1 B(2) C(3)

2 D(2) E(3) E(1) F(2)

Figure 6.13
© 2009 Prentice-Hall, Inc. 6 – 91
 Material Structure Tree

 It is clear from the three that the demand for B, C,

D, E, and F is completely dependent on the
demand for A
 The material structure tree has three levels: 0, 1,
and 2
 Items above a level are called parents
 Items below any level are called components
 The number in parenthesis beside each item
shows how many are required to make the item
above it

© 2009 Prentice-Hall, Inc. 6 – 92

 Material Structure Tree
 We can use the material structure tree and the
demand for Item A to compute demands for the
other items

Part B: 2  number of A’s  2  50  100

Part C: 3  number of A’s  3  50  150
Part D: 2  number of B’s  2 100  200
Part E: 3  number of B’s + 1  number of C’s
 3  100 + 1 150  450
Part F: 2  number of C’s  2  150  300

© 2009 Prentice-Hall, Inc. 6 – 93

 Gross and Net Material
Requirements Plan
 Once the materials structure tree is done, we
construct a gross material requirements plan
 This is a time schedule that shows when an item
must be ordered when there is no inventory on
hand, or
 When the production of an item must be started in
order to satisfy the demand for the finished
product at a particular date
 We need lead times for each of the items

Item A – 1 week Item D – 1 week

Item B – 2 weeks Item E – 2 weeks
Item C – 1 week Item F – 3 weeks
© 2009 Prentice-Hall, Inc. 6 – 94
 Gross Material Requirements Plan
Week
1 2 3 4 5 6
Required Date 50
A Lead Time = 1 Week
Order Release 50

Required Date 100

B Lead Time = 2 Weeks
Order Release 100

Required Date 150

C Lead Time = 1 Week
Order Release 150

Required Date 200

D Lead Time = 1 Week
Order Release 200

Required Date 300 150

E Lead Time = 2 Weeks
Order Release 300 150

Required Date 300

F Lead Time = 3 Weeks
Order Release 300
Figure 6.14 © 2009 Prentice-Hall, Inc. 6 – 95
 Net Material Requirements Plan

 A net material requirements plan can be

constructed from the gross materials
requirements plan and on-hand inventory
information

ITEM ON-HAND INVENTORY

A 10
B 15
C 20
D 10
E 10
F 5

Table 6.11
© 2009 Prentice-Hall, Inc. 6 – 96
 Net Material Requirements Plan

 Using this data we can construct a plan

that includes
 Gross requirements
 On-hand inventory
 Net requirements
 Planned-order receipts
 Planned-order releases
 The net requirements plan is constructed
like the gross requirements plan

© 2009 Prentice-Hall, Inc. 6 – 97

 Net Material Requirements Plan

Week Lead
Item 1 2 3 4 5 6 Time

A Gross 50 1
On-Hand 10 10
Net 40
Order Receipt 40
Order Release 40
B Gross 80A 2
On-Hand 15 15
Net 65
Order Receipt 65
Order Release 65

Figure 6.15(a)

© 2009 Prentice-Hall, Inc. 6 – 98

 Net Material Requirements Plan

Week Lead
Item 1 2 3 4 5 6 Time

C Gross 120A 1
On-Hand 20 10
Net 100
Order Receipt 100
Order Release 100
D Gross 130B 1
On-Hand 10 10
Net 120
Order Receipt 120
Order Release 120

Figure 6.15(b)

© 2009 Prentice-Hall, Inc. 6 – 99

 Net Material Requirements Plan

Week Lead
Item 1 2 3 4 5 6 Time

E Gross 195B 100C 2

On-Hand 10 10 0
Net 185 100
Order Receipt 185 100
Order Release 185 100
F Gross 200C 3
On-Hand 5 5
Net 195
Order Receipt 195
Order Release 195

Figure 6.15(c)

© 2009 Prentice-Hall, Inc. 6 – 100

 Two or More End Products

 Most manufacturing companies have more than

one end item
 In this example, the second product is AA and it
has the following material structure tree
AA

D(3) F(2)

 If we require 10 units of AA, the gross

requirements for parts D and F can be computed
Part D: 3  number of AA’s  3  10  30
Part F: 2  number of AA’s  2  10  20

© 2009 Prentice-Hall, Inc. 6 – 101

 Two or More End Products

 The lead time for AA is one week

 The gross requirement for AA is 10 units in week 6
and there are no units on hand
 This new product can be added to the MRP
process
 The addition of AA will only change the MRP
schedules for the parts contained in AA

 MRP can also schedule spare parts and

components
 These have to be included as gross requirements

© 2009 Prentice-Hall, Inc. 6 – 102

 Two or More End Products

Week Lead
Item 1 2 3 4 5 6 Time

AA Gross 10 1
On-Hand 0 0
Net 10
Order Receipt 10
Order Release 10

Figure 6.16(a)

© 2009 Prentice-Hall, Inc. 6 – 103

 Two or More End Products

Week Lead
Item 1 2 3 4 5 6 Time

D Gross 130B 30AA 1

On-Hand 10 10 0
Net 120 30
Order Receipt 120 30
Order Release 120 30
F Gross 200C 20AA 3
On-Hand 5 5 0
Net 195 20
Order Receipt 195 20
Order Release 195 20

Figure 6.16(b)

© 2009 Prentice-Hall, Inc. 6 – 104

 Just-in-Time Inventory Control

 To achieve greater efficiency in the

production process, organizations have
tried to have less in-process inventory on
hand
 This is known as JIT inventory
 The inventory arrives just in time to be
used during the manufacturing process
 One technique of implementing JIT is a
manual procedure called kanban

© 2009 Prentice-Hall, Inc. 6 – 105

 Just-in-Time Inventory Control

 Kanban in Japanese means “card”

 With a dual-card kanban system, there is a
conveyance kanban, or C-kanban, and a
production kanban, or P-kanban
 Kanban systems are quite simple, but they
require considerable discipline
 As there is little inventory to cover
variability, the schedule must be followed
exactly

© 2009 Prentice-Hall, Inc. 6 – 106

 4 Steps of Kanban
1. A user takes a container of parts or inventory
along with its C-kanban to his or her work area
When there are no more parts or the container is
empty, the user returns the container along with
the C-kanban to the producer area
2. At the producer area, there is a full container of
parts along with a P-kanban
The user detaches the P-kanban from the full
container and takes the container and the C-
kanban back to his or her area for immediate
use

© 2009 Prentice-Hall, Inc. 6 – 107

 4 Steps of Kanban
3. The detached P-kanban goes back to the
producer area along with the empty container
The P-kanban is a signal that new parts are to be
manufactured or that new parts are to be placed
in the container and is attached to the container
when it is filled
4. This process repeats itself during the typical
workday

© 2009 Prentice-Hall, Inc. 6 – 108

 The Kanban System

P-kanban C-kanban
and and
Container Container

4 1

Producer Storage User

Area Area Area

3 2

Figure 6.17

© 2009 Prentice-Hall, Inc. 6 – 109

 Enterprise Resource Planning

 MRP has evolved to include not only the materials

required in production, but also the labor hours,
material cost, and other resources related to
production
 In this approach the term MRP II is often used and
the word resource replaces the word requirements
 As this concept evolved and sophisticated
software was developed, these systems became
known as enterprise resource planning (ERP)
systems

© 2009 Prentice-Hall, Inc. 6 – 110

 Enterprise Resource Planning

 The objective of an ERP System is to reduce costs

by integrating all of the operations of a firm
 Starts with the supplier of materials needed and
flows through the organization to include
invoicing the customer of the final product
 Data are entered only once into a database where
it can be quickly and easily accessed by anyone in
the organization
 Benefits include
 Reduced transaction costs
 Increased speed and accuracy of information
 Almost all areas of the firm benefit

© 2009 Prentice-Hall, Inc. 6 – 111

 Enterprise Resource Planning

 There are drawbacks

 The software is expensive to buy and costly to
customize
 Small systems can cost hundreds of thousands of
dollars
 Large systems can cost hundreds of millions
 The implementation of an ERP system may require
a company to change its normal operations
 Employees are often resistant to change
 Training employees on the use of the new
software can be expensive

© 2009 Prentice-Hall, Inc. 6 – 112