1

Inventory Models
Chapter 8
2
8.1 Overview of Inventory Issues
Proper control of inventory is crucial to the success of an
enterprise.
Typical inventory problems include:
Basic inventory Planned shortage
Quantity discount Periodic review
Production lot size Single period
Inventory models are often used to develop an optimal
inventory policy, consisting of:
An order quantity, denoted Q.
A reorder point, denoted R.
3
Inventory analyses can be thought of as cost-control
techniques.
Categories of costs in inventory models:
Holding (carrying costs)
Order/ Setup costs
Customer satisfaction costs
Procurement/Manufacturing costs
Type of Costs in Inventory Models
4
Holding Costs (Carrying costs):
These costs depend on the order size
Cost of capital
Storage space rental cost
Costs of utilities
Labor
Insurance
Security
Theft and breakage
Deterioration or Obsolescence



C
h
= Annual holding cost per unit
in inventory
H = Annual holding cost rate
C = Unit cost of an item
C
h
= H * C
Type of Costs in Inventory Models
5
Order/Setup Costs
These costs are independent of the order size.
Order costs are incurred when purchasing a good
from a supplier. They include costs such as
Telephone
Order checking
Labor
Transportation
Setup costs are incurred when producing goods for
sale to others. They can include costs of
Cleaning machines
Calibrating equipment
Training staff
Type of Costs in Inventory Models
C
o
= Order cost
or setup cost
6
Customer Satisfaction Costs
Measure the degree to which a
customer is satisfied.
Unsatisfied customers may:
Switch to the competition (lost sales).
Wait until an order is supplied.
When customers are willing to wait
there are two types of costs
incurred:
Type of Costs in Inventory Models
C
b
= Fixed administrative
costs of an out of stock
item ($/stockout unit).
C
s
= Annualized cost of a
customer awaiting an
out of stock item
($/stockout unit per
year).
7
Procurement/Manufacturing Cost
Represents the unit purchase cost
(including transportation) in case of a
purchase.
Unit production cost in case of in-
house manufacturing.
Type of Costs in Inventory Models
C = Unit purchase or
manufacturing cost.
8
Demand is a key component affecting an inventory
policy.
Projected demand patterns determine how an
inventory problem is modeled.
Typical demand patterns are:
Constant over time (deterministic inventory models)
Changing but known over time (dynamic models)
Variable (randomly) over time (probabilistic models)
Demand in Inventory Models
D = Demand rate (usually per year)
9
Inventory can be classified in various ways:
By Process By Importance By Shelf Life
Raw materials Perishable
Work in progress A, B, C Nonperishable
Finished goods
By Process By Importance By Shelf Life
Raw materials Perishable
Work in progress A, B, C Nonperishable
Finished goods
Used typically by accountants
at manufacturing firms.
Enables management to track
the production process.
Items are classified by
their relative importance
in terms of the firms
capital needs.
Management of items
with short shelf life and
long shelf life is very
different
Inventory Classifications
10
Two types of review systems are used:
Continuous review systems.
The system is continuously monitored.
A new order is placed when the inventory reaches a critical
point.
Periodic review systems.
The inventory position is investigated on a regular basis.
An order is placed only at these times.
Review Systems
11
The item has a sufficiently long shelf life.
The item is monitored using a continuous review
system.
All the cost parameters remain constant forever
(over an infinite time horizon).
A complete order is received in one batch.
8.2 Economic Order Quantity Model -
Assumptions
Demand occurs at a known and reasonably
constant rate.
12
The constant environment described by the EOQ
assumptions leads to the following observation:
The optimal EOQ policy consists of same-size orders.
Q Q Q
The EOQ Model
Inventory profile
This observation results in the following inventory profile :
13
Q Q Q
Total Annual
Inventory Costs
=
Total Annual
Holding Costs
Total Annual
ordering Costs
Total Annual
procurement Costs
+ +
TC(Q) = (Q/2)C
h
+ (D/Q)C
o
+ DC
C
h
The optimal order Size
2DC
o
Q
*
=
Cost Equation for the EOQ Model
14
Constructing the total annual variable cost curve
Add the two curves to one another
Total annual holding and
ordering costs
Q
TV(Q)
Q
*
The optimal order size
o
*
*
*
*
*
TV(Q) and Q
*

15
The curve is reasonably flat around Q
*
.

Q
*
Deviations from the optimal order size
cause only small increase in the total cost.
Sensitivity Analysis in EOQ models
16
The cycle time, T, represents the time that
elapses between the placement of orders.



Note, if the cycle time is greater than the shelf
life, items will go bad, and the model must be
modified.
T = Q/D
Cycle Time
17
To find the number of orders per years take the
reciprocal of the cycle time
N = D/Q
Example: The demand for a product is 1000 units per year.
The order size is 250 units under an EOQ policy.
How many orders are placed per year? N = 1000/250 = 4 orders.
How often orders need to be placed (what is the cycle time)?
T = 250/1000 = years. {Note: the four orders are equally spaced}.
Number of Orders per Year
18
In reality lead time always exists, and must be
accounted for when deciding when to place an
order.
The reorder point, R, is the inventory position
when an order is placed.
R is calculated by

L and D must be expressed in the same time unit.

R = L D
Lead Time and the Reorder Point
19
L
Place the order
now
Reorder
Point
R = Inventory at hand at the beginning of lead time
Lead Time and the Reorder Point
Graphical demonstration: Short Lead Time
20
Outstanding order
Place the order
now
R = inventory at hand
at the beginning of lead time +
one outstanding order
= demand during lead time
= LD
Inventory
at
hand
L
Lead Time and the Reorder Point
Graphical demonstration: Long Lead Time
21
Safety stocks act as buffers to handle:
Higher than average lead time demand.
Longer than expected lead time.
With the inclusion of safety stock (SS), R is calculated by



The size of the safety stock is based on having a desired
service level.

R = LD + SS
Safety stock
22
L
Place the order
now
Reorder
Point
R = LD
Safety stock
Planned
situation
Actual
situation
23
L
R = LD
Safety stock
Actual
situation
+ SS
Reorder
Point
Place the order
now
SS=Safety stock
?
The safety stock
prevents excessive
shortages.
LD
24
Inventory Costs
Including safety stock
Total Annual
Inventory Costs
=
Total Annual
Holding Costs
Total Annual
ordering Costs
Total Annual
procurement Costs
+ +
TC(Q) = (Q/2)C
h
+ (D/Q)C
o
+ DC + C
h
SS
Safety stock
holding cost
25
ALLEN APPLIANCE COMPANY (AAC)
AAC wholesales small appliances.
AAC currently orders 600 units of the Citron brand
juicer each time inventory drops to 205 units.
Management wishes to determine an optimal
ordering policy for the Citron brand juicer
26
Sales of Juicers over the last 10 weeks
Week 1 2 3 4 5
Sales 105 115 125 120 125
Week 6 7 8 9 10
Sales 120 135 115 110 130
Data
C
o
= $12 ($8 for placing an order) + (20 min. to check)($12 per hr)
C
h
= $1.40 [HC = (14%)($10)]
C = $10.
H = 14% (10% ann. interest rate) + (4% miscellaneous)
D = demand information of the last 10 weeks was collected:
ALLEN APPLIANCE COMPANY (AAC)
27
Data
The constant demand rate seems to be a good
assumption.
Annual demand = (120/week)(52weeks) = 6240 juicers.
ALLEN APPLIANCE COMPANY (AAC)
28
Current ordering policy calls for Q = 600 juicers.
TV( 600) = (600 / 2)($1.40) + (6240 / 600)($12) = $544.80

The EOQ policy calls for orders of size




AAC Solution:
EOQ and Total Variable Cost
Savings of 16%
\
2(6240)(12)
1.40
= 327.065 327
=
Q
*

TV(327) = (327 / 2)($1.40) + (6240 / 327) ( $12) = $457.89
29
TC(327) = 457.89 + 6240($10) + (13)($1.40) = $62,876.09
Under the current ordering policy AAC holds 13 units safety
stock (how come? Observe):

AAC is open 5 day a week.
The average daily demand = 120/week)/5 = 24 juicers.
Lead time is 8 days. Lead time demand is (8)(24) = 192 juicers.
Reorder point without Safety stock = LD = 192.
Current policy: R = 205.
Safety stock = 205 192 = 13.

For safety stock of 13 juicers the total cost is
TV(327) + Procurement + Safety stock
cost holding cost
AAC Solution:
Reorder Point and Total Cost
30
Changing the order size
Suppose juicers must be ordered in increments of 100 (order 300 or 400)
AAC will order Q = 300 juicers in each order.
There will be a total variable cost increase of $1.71.
This is less than 0.5% increase in variable costs.

Changes in input parameters
Suppose there is a 20% increase in demand. D=7500 juicers.
The new optimal order quantity is Q
*
= 359.
The new variable total cost = TV(359) = $502
If AAC still orders Q = 327, its total variable costs becomes
TV(327) = (327/2)($1.40) + (7500/327)($12) = $504.13
Only 0.4%
increase
AAC Solution:
Sensitivity of the EOQ Results
31
For an order size of 327 juicers we have:
T = (327/ 6240) = 0.0524 year.
= 0.0524(52)(5) = 14 days.


This is useful information because:
Shelf life may be a problem.
Coordinating orders with other items might be desirable.
~
AAC Solution:
Cycle Time
working days per week
32
AAC Excel Spreadsheet
=SQRT(2*$B$10*$
B$14/$B$13)
=1/E11
Copy to cell H12
=E10/B10
Copy to cell
H11
=$B$10*$B$11+E14+$B$13*B16
Copy to Cell H15
=(E10/2)*$B$13+($B$10/E10)*$B$14
Copy to cell H14
=$B$15*$B$10+$B$16-
INT(($B$15*$B$10+$B$16)/E10)*E10
Copy to cell H13
33
Service Levels and
Safety Stocks
34
8.3 Determining Safety Stock Levels
Businesses incorporate safety stock requirements
when determining reorder points.
A possible approach to determining safety stock
levels is by specifying desired service level .
35
The unit service level
The percentage of demands
that are filled without
incurring any delay.
Applied when the percentage
of unsatisfied demand
should be under control.
Two Types of Service Level
The cycle service level
The probability of not
incurring a stockout during
an inventory cycle.
Applied when the likelihood
of a stockout, and not its
magnitude, is important for
the firm.
Service levels can be viewed in two ways.
36

In many cases short run demand is variable even
though long run demand is assumed constant.

Therefore, stockout events during lead time may
occur unexpectedly in each cycle.

Stockouts occur only if demand during lead time is
greater than the reorder point.

The Cycle Service Level Approach
37
To determine the reorder point we need to know:
The lead time demand distribution.
The required service level.

In many cases lead time demand is approximately
normally distributed. For the normal distribution case
the reorder point is calculated by

The Cycle Service Level Approach
R =
L
+ z
o
o
L
1 o = service level
38
=192
P(D
L
> R) = P(Z > (R
L
)/o
L
) = o. Since
P(Z > Z
o
) = o, we have Z
o
= (R
L
)/o
L,
which gives
The Cycle Service Level Approach
P(D
L
>R) = o
Service level =
P(D
L
<R) =
1 o
R
R =
L
+ z
o
o
L
39
Assume that lead time demand is normally
distributed.

Estimation of the normal distribution parameters:

Estimation of the mean weekly demand =
ten weeks average demand = 120 juicers per week.

Estimation of the variance of the weekly demand =
Sample variance = 83.33 juicers
2
.
AAC -
Cycle Service Level Approach
40
To find
L
and

o
L
the parameters (per week) and o
(per week) must be adjusted since the lead time is
longer than one week.

Lead time is 8 days =(8/5) weeks = 1.6 weeks.

Estimates for the lead time mean demand and
variance of demand

L
~ (1.6)(120) = 192; o
2
L
~ (1.6)(83.33) = 133.33
AAC -
Cycle Service Level Approach
41
Let us use the current reorder point of 205 juicers.

205 = 192 + z (11.55) z = 1.13



From the normal distribution table we have that a reorder
point of 205 juicers results in an 87% cycle service level.

133 33 .
AAC -
Service Level for a given Reorder Point
42
Management wants to improve the cycle service
level to 99%.
The z value corresponding to 1% right hand tail is
2.33.
R = 192 + 2.33(11.55) = 219 juicers.
AAC
Reorder Point for a given Service Level
43
AAC is willing to run out of stock an average of at
most one cycle per year with an order quantity of
327 juicers.

What is the equivalent service level for this
strategy?
AAC
Acceptable Number of Stockouts per Year
44
AAC
Acceptable Number of Stockouts per Year
There will be an average of
6240/327 = 19.08 lead times per year.
The likelihood of stockouts = 1/19 = 0.0524.

This translates into a service level of 94.76%

45
When lead time demand follows a normal distribution
service level can be calculated as follows:
Determine the value of z that satisfy the equation

L(z) = oQ
*
/ o
L



Solve for R using the equation
R =
L
+ zo
L
The Unit Service Level Approach
46
=NORMDIST(B8,B5,B6,TRUE)
AAC
Cycle Service Level (Excel spreadsheet)
=NORMINV(B7,B5,B6)
47
Quantity Discounts are Common Practice in Business
By offering discounts buyers are encouraged to increase
their order sizes, thus reducing the sellers holding costs.

Quantity discounts reflect the savings inherent in large
orders.

With quantity discounts sellers can reward their biggest customers
without violating the Robinson - Patman Act.
8.4 EOQ Models with Quantity Discounts
48
Quantity Discount Schedule
This is a list of per unit discounts and their corresponding
purchase volumes.
Normally, the price per unit declines as the order quantity
increases.
The order quantity at which the unit price changes is called a
break point.
There are two main discount plans:
All unit schedules - the price paid for all the units purchased is based on
the total purchase.
Incremental schedules - The price discount is based only on the
additional units ordered beyond each break point.
8.4 EOQ Models with Quantity Discounts
49
To determine the optimal order quantity, the total
purchase cost must be included

TC(Q) = (Q/2)C
h
+ (D/Q)C
o
+ DC
i
+ C
h
SS

C
i
represents the unit cost at the i
th
pricing level.
All Units Discount Schedule
50
AAC - All Units Quantity Discounts
Quantity Discount
Schedule
1-299 $10.00
300-599 $9.75
600-999 $9.40
1000-4999 $9.50
5000 $9.00
Quantity Discount
Schedule
1-299 $10.00
300-599 $9.75
600-999 $9.40
1000-4999 $9.50
5000 $9.00
>
AAC is offering all units quantity discounts to its
customers.
Data
51
Should AAC increase its regular order of
327 juicers, to take advantage of the discount?
52
AAC All units discount procedure
Step 1: Find the optimal order Q
i
*
for each discount level i.
Use the formula
Step 2: For each discount level i modify Q
i
*
as follows
If Q
i
*
is lower than the smallest quantity that qualifies for the i
th

discount, increase Q
i
*
to that level.
If Q
i
*
is greater than the largest quantity that qualifies for the i
th

discount, eliminate this level from further consideration.
Step 3: Substitute the modified Q
*
i
value in the total cost
formula TC(Q
*
i
).
Step 4: Select the Q

i
*
that minimizes TC(Q
i
*
)
Q DC C o h * ( ) / = 2
53
Step 1: Find the optimal order quantity Q
i
*
for each
discount level i based on the EOQ formula
Lowest cost order size per discount level
Discount Qualifying Price
level order per unit Q*
0 1-299 10.00 327
1 300-599 9.75 331
2 600-999 9.50 336
3 1000-4999 9.40 337
4 5000 9.00 345 >
AAC All units discount procedure
54
Step 2 : Modify Q
i
*

Modified Q* and total Cost
Qualified Price Modified Total
Urder per Unit Q* Q* Cost
1-299 10.00 300 **** ****
300-599 9.75 331 331 61,292.13
600-999 9.50 336 600 59,803.80
1000-4999 9.40 337 1000 59,388.88
5000 9.00 345 5000 59,324.98
Modified Q* and total Cost
Qualified Price Modified Total
Urder per Unit Q* Q* Cost
1-299 10.00 300 **** ****
300-599 9.75 331 331 61,292.13
600-999 9.50 336 600 59,803.80
1000-4999 9.40 337 1000 59,388.88
5000 9.00 345 5000 59,324.98
>
1 299
Q
1
*

300
$10/unit
599
331
Q
2
*

$9.75/unit
999 600
Q
3
*

336
$9.50
AAC All Units Discount Procedure
55
Step 2 : Modify Q
i
*

Modified Q* and total Cost
Qualified Price Modified Total
Urder per Unit Q* Q* Cost
1-299 10.00 300 **** ****
300-599 9.75 331 331 61,292.13
600-999 9.50 336 600 59,803.80
1000-4999 9.40 337 1000 59,388.88
5000 9.00 345 5000 59,324.98
Modified Q* and total Cost
Qualified Price Modified Total
Urder per Unit Q* Q* Cost
1-299 10.00 300 **** ****
300-599 9.75 331 331 61,292.13
600-999 9.50 336 600 59,803.80
1000-4999 9.40 337 1000 59,388.88
5000 9.00 345 5000 59,324.98
1 299
Q
1
*

300
$10/unit
331
Q
2
*

999 600
Q
3
*

336
$9.50
AAC All Units Discount Procedure
Q
3
*

Q
3
*

Q
3
*

Q
3
*

Q
3
*

Q
3
*

Q
3
*

56
Step 3: Substitute Q
I
*
in the total cost function






Step 4
Modified Q* and total Cost
Qualified Price Modified Total
Urder per Unit Q* Q* Cost
1-299 10.00 300 **** ****
300-599 9.75 331 331 61,292.13
600-999 9.50 336 600 59,803.80
1000-4999 9.40 337 1000 59,388.88
5000 9.00 345 5000 59,324.98
Modified Q* and total Cost
Qualified Price Modified Total
Urder per Unit Q* Q* Cost
1-299 10.00 300 **** ****
300-599 9.75 331 331 61,292.13
600-999 9.50 336 600 59,803.80
1000-4999 9.40 337 1000 59,388.88
5000 9.00 345 5000 59,324.98 >
AAC should order 5000 juicers
AAC All Units Discount Procedure
57
Calculation of Optimal Inventory Policy Under All-Units Quantity Discounts
OPTIMAL
INPUTS Values OUTPUTS Values
Annual Demand, D = 6240.00 Order quantity, Q* = 5000
Per Unit Cost, C = 10.00 Cycle Time (in years), T = 0.801282051
Annual Holding Cost Rate, H = 0.14 # of Cycles Per Year, N = 1.248
Annual Holding Cost Per Unit, Ch = 1.40 Reorder Point, R = 205.0000
Order Cost, Co = 12.00 Total Annual Cost, TC(Q*) = 59341.36
Lead Time (in years), L = 0.03077
Safety Stock, SS = 13.00
DISCOUNTS
Level Breakpoint Discount Price Q* TC(Q*) Modified Q*
0 1 10.00 327 62876.09 327
1 300 9.75 331 61309.88 331
2 600 9.50 336 59821.09 600
3 1000 9.40 337 59405.99 1000
4 5000 9.00 345 59341.36 5000
5
6
7
8
AAC All Units Discount Excel Worksheet
58
Demand rate is constant.

Production rate is larger than demand rate.
The production lot is not received instantaneously (at an
infinite rate), because production rate is finite.

There is only one product to be scheduled.

The rest of the EOQ assumptions stay in place.
8.4 Production Lot Size Model -
Assumptions
59
The optimal production lot size policy orders the
same amount each time.
This observation results in the inventory profile
below:

Production Lot Size Model
Inventory profile
60
Production
Lot Size = Q = PT
1

The inventory increases
at a net rate of P - D
The production increases the
inventory at a rate of P.
The demand decreases the
inventory at a rate of D.
Production
time
T
1

Demand accumulation
during production run
Demand accumulation
during production run = DT
1

Maximum inventory = (P D)T
1
= (P D)(Q/P) = Q(1 D/P)
Maximum inventory
Production Lot Size Model
Understanding the inventory profile
61


The parameters of the total variable costs function are
similar to those used in the EOQ model.

Instead of ordering cost, we have here a fixed setup
cost per production run (C
o
).

In addition, we need to incorporate the annual
production rate (P) in the model.
Production Lot Size Model
Total Variable Cost
62
TV(Q) = (Q/2)(1 - D/P)C
h
+ (D/Q)C
o

P is the annual production rate
C
h
(1-D/P)
The Optimal Order Size
Q
*
=
2DC
o
The average inventory
Production Lot Size Model
Total Variable Cost
63
Cycle time T = Q / D.

Length of a production run T
1
= Q / P.

Time when machines are not busy producing the
product T
2
= T - T
1
= Q(1/D - 1/P).

Average inventory = (Q/2)(1-D/P).
Production Lot Size Model
Useful relationships
64
FARAH COSMETICS COMPANY
Farah needs to determine optimal production lot
size for its most popular shade of lipstick.
Data
The factory operates 7 days a week, 24 hours a day.
Production rate is 1000 tubes per hour.
It takes 30 minutes to prepare the machinery for production.
It costs $150 to setup the line.
Demand is 980 dozen tubes per week.
Unit production cost is $.50
Annual holding cost rate is 40%.
65
Input for the total variable cost function
D = 613,200 per year [(980 dozen/week (12)/ 7](365)
C
h
= 0.4(0.5) = $0.20 per tube per year.
C
o
= $150
P = (1000)(24)(365) = 8,760,000 per year.

Dozens
FARAH COSMETICS COMPANY
Solution
66
Current Policy
Currently, Farah produces in lots of 84,000 tubes.

T = (84,000 tubes per run)/(613,200 tubes per year)= 0.137 years
(about 50 days).

T
1
= (84,000 tubes per lot)/(8,760,000 tubes per year)= 0.0096 years
(about 3.5 days).

T
2
= 0.137 - 0.0096 = 0.1274 years (about 46.5 days).

TV(Q = 84,000) = (84,000/2) {1-(613,200/8,760,000)}(0.2)
+ 613,200/84,000)(150) = $8907.
FARAH COSMETICS COMPANY
Solution
67
The Optimal Policy

Using the input data we find



TV(Q
*
= 31,499) = (31,499/2) [1-(613,200/8,760,000)](0.2) +
(613,200/31,499)(150) = $5,850.
The optimal order size
(0.2)(1-613,200/8760,000)
Q
*
=
2(613,200)(150)
= 31,499
FARAH COSMETICS COMPANY
Solution
68
FARAH COSMETICS COMPANY
Production Lot Size Template (Excel)
69
8.5 Planned Shortage Model
When an item is out of stock, customers may:
Go somewhere else (lost sales).
Place their order and wait (backordering).

In this model we consider the backordering case.

All the other EOQ assumptions are in place.
70

The parameters of the total variable costs function are
similar to those used in the EOQ model.

In addition, we need to incorporate the shortage costs in
the model.
Backorder cost per unit per year (loss of goodwill cost) - C
s
.
Reflects future reduction in profitability.
Can be estimated from market surveys and focus groups.
Backorder administrative cost per unit - C
b
.
Reflects additional work needed to take care of the backorder.
Planned Shortage Model
the Total Variable Cost Equation
71
Planned Shortage Model
the Total Variable Cost Equation
The Annual holding cost =
C
h
[T
1
/T](Average inventory) =
C
h
[T
1
/T] (Q-S)/2

The Annual shortage cost =
C
b
(number of backorders per year) +
C
s
(T
2
/T)(Average number of backorders).

To calculate the annual holding cost and
shortage cost we need to find
The proportion of time inventory is carried, (T
1
/T)
The proportion of time demand is backordered, (T
2
/T).
T
1
T
2
T
72
S
Q - S
Q
T
1
T
2
S
T
Average inventory = (Q - S) / 2
Average shortage = S / 2
Proportion of time
inventory exists
= T
1
/T
T
1
T
Q
Proportion of time
shortage exists
= T
2
/T
Finding T
1
/ T and T
2
/ T
= (Q - S) / Q
= S / Q
73
Annual holding cost:
C
h
[T
1
/T](Q-S)/2 = C
h
[(Q-S) /Q](Q-S)/2
= C
h
(Q-S)
2
/2Q

Annual shortage cost:
C
b
(Units in short per year) +
C
s
[T
2
/T](Average number of backorders) =
C
b
(S)(D/Q) + C
s
S
2
/2Q
Planned Shortage Model
The Total Variable Cost Equation
74
The total annual variable cost equation





The optimal solution to this problem is obtained under the
following conditions
C
s
> 0 ;
C
b
< \/ 2C
o
C
h
/ D
TV(Q,S) =
(Q -S)
2
2Q
C
h
+
D
Q
(C
o
+ SC
b
) +
S
2

2Q
C
S
Holding
costs
Time dependent
backorder costs
Time independent
backorder costs
Ordering
costs
Planned Shortage Model
The Total Variable Cost Equation
75
The Optimal Backorder level
S
*
=
Q
*
C
h
- DC
b

C
h
+ C
s
Reorder Point
R = L D - S
*
Planned Shortage Model
The Optimal Inventory Policy
The Optimal Order Size
C
h
(DC
b
)
2

C
h
C
s
2DC
o

Q
*
=
C
h
+ C
s

C
s
x
76
SCANLON PLUMBING CORPORATION
Scanlon distributes a portable sauna from Sweden.
Data
A sauna costs Scanlon $2400.
Annual holding cost per unit $525.
Fixed ordering cost $1250 (fairly high, due to costly transportation).
Lead time is 4 weeks.
Demand is 15 saunas per week on the average.
77

Scanlon estimates a $20 goodwill cost for each week a
customer who orders a sauna has to wait for delivery.
Administrative backordrer cost is $10.

Management wishes to know:
The optimal order quantity.
The optimal number of backorders.
Backorder costs
SCANLON PLUMBING CORPORATION
78
SCANLON PLUMBING
Solution
Input for the total variable cost function

D = 780 saunas [(15)(52)]
C
o
= $1,250
C
h
= $525
C
s
= $1,040
C
b
= $10
79
x
(780)(10)
2

(525)(1040)
525
2(780)(1250) 525+1040
1040
Q
*
=
74 ~
The optimal policy

R = (4 / 52)(780) 20 = 40
_
S
*
=
(74)(525) (780)(10)
525 + 1040
20
~
SCANLON PLUMBING
Solution
80
SCANLON PLUMBING
Spreadsheet Solution
Calculation of Optimal Inventory Policy for a Planned Shortage Model
OPTIMAL ASSIGNED
INPUTS Values OUTPUTS Values OUTPUTS Values
Annual Demand, D = 780.00 Order Quantity, Q* = 74.01 Q = 74.00
Per Unit Cost, C = 2400.00 Backorder Level, S* = 19.84 S = 20.00
Annual Holding Cost Rate, H = 0.22 Cycle Time (in years), T = 0.0949 T = 0.0949
Annual Holding Cost Per Unit, Ch = 525.00 # of Cycles Per Year, N = 10.5388 N = 10.5405
Order Cost, Co = 1250.00 Reorder Point, R = 40.1531 R = 39.9976
Annual Backorder Cost, Cs = 1040.00 Total Annual Variable Cost, TV(Q*) = 28438.24 TV(Q) = 28438.51
Fixed Admin. Backorder Cost, Cb = 10.00 Total Annual Cost, TC(Q*) = 1900438.24 TC(Q) = 1900438.51
Lead Time (in years), L = 0.07692 % of Customers Backordered = 26.81 % Back. = 27.03
81
8.7 Review Systems Continuous
Review

(R, Q) Policies
The EOQ, production lot size, and planned shortage
models assume that
inventory levels are continuously monitored
Items are sold one at a time.
82

(R, Q) Policies
The above models call for order point (R) order
quantity (Q) inventory policies.
Such policies can be implemented by
A point-of-sale computerized system.
The two-bin system.
8.7 Review Systems Continuous
Review
83
(R, M) policies
When items are not necessarily sold one at a time, the
reorder point might be missed, and out of stock
situations might occur more frequently.
The order to level (R, M) policy may be implemented in
this situation.
Continuous Review Systems
84
(R,M) policies
The R, M policy replenishes inventory up to a pre-
determined level M.
Continuous Review Systems
Order Q = Q
*
+ (R I) = (M SS) + (R I) each
time the inventory falls to the reorder point R or below.
(Order size may vary from one cycle to another).
85
It may be difficult or impossible to adopt a
continuous review system, because of:
The high price of a computerized system.
Lack of space to adopt the two-bin system.
Operations inefficiency when ordering different items from
the same vendor separately.
The periodic review system may be found more suitable
for these situations.
Periodic Review Systems
86
Under this system the inventory position for each
item is observed periodically.

Orders for different items can be better
coordinated periodically.
Periodic Review Systems
87
(T,M) Policies
In a replenishment cycle policy (T, M), the
inventory position is reviewed every T time units.
An order is placed to bring the inventory level back up
to a maximum inventory level M.
M is determined by
Forecasting the number of units demanded during the
review period T.
Adding the desired safety stock to the forecasted demand.
Periodic Review Systems
88
T =Review period
L = Lead time
SS= Safety stock
Q = Inventory position
D = Annual demand
I = Inventory position
Periodic Review Systems
Calculation of the replenishment level and order
size
Q = M + LD I
M = TD + SS
89
Every three weeks AAC receives deliveries of
different products from Citron.
Lead time is eight days for ordering Citrons
juicers.
AAC is now reviewing its juicer inventory and
finds 210 in stock.
How many juicers should AAC order for a safety
stock of 30 juicers?
AAC operates a (T, M) policy
90
Data
Review period T = 3 weeks = 3/52 = .05769 years,
Lead time = L = 8 days = 8/260 = .03077 years,
Demand D = 6240 juicers per year,
Safety stock SS = 30 juicers,
Inventory position I = 210 juicers
AAC operates a (T, M) policy Solution
AAC operates
260 days a year.
(5)(52) = 260.
91
Review period demand = TD = ( 3/52)(6240) = 360
juicers,

M = TD + SS = 360 + 30 = 390 juicers,

Q = M + LD I = 390 + .03077(6240) - 210 = 372
juicers.
AAC operates a (T, M) policy Solution
92
Review
point
Review
point
AAC operates a (T, M) policy Solution
T
SS SS SS
Inventory position
Order
Order
Replenishment level
Inventory position
L
Notice: I + Q is designed to satisfy the demand within an interval of T + L.
To obtain the replenishment level add SS to I + Q.
M = maximum inventory
L
93
Demand is stochastic
with a known distribution.

Shelf life of the item is
limited.
Inventory is saleable only
within a single time
period.
Inventory is delivered
only once during a time
period.
8.8 Single Period Inventory Model -
Assumptions
At the end of each
period, unsold inventory
is disposed of for some
salvage.
The salvage value is less
than the cost per item.
Unsatisfied demand may
result in shortage costs.

94
To find an optimal order quantity we need to balance the
expected cost of over-ordering and under ordering.

Expected Profit = E(Profit when Demand=X)Prob(Demand=X)

x
The expected profit is a function of the order size, the random
demand, and the various costs.
The Expected Profit Function
95
Developing an expression for EP(Q)
Notation
p = per unit selling price of the good.
c = per unit cost of the good.
s = per unit salvage value of unsold good.
K = fixed purchasing costs
Q = order quantity.
EP(Q) = Expected Profit if Q units are ordered.

Scenarios
Demand X is less than the order quantity (X < Q).
Demand X is greater than or equal to the order quantity (X >Q.
The Expected Profit Function
96
Scenario 1: Demand X is less than the units
stocked, Q.

Scenario 2: Demand X is greater than or equal to the
units stocked.
Profit = pX + s(Q - X) - cQ - K
Profit = pQ - g(X - Q) - cQ - K
EP(Q) = [pX+s(Q - X) - cQ - K]P(X) + [pQ - g(X - Q) - cQ - K]P(X)
X Q <

X Q >

The Expected Profit Function

97
To maximize the expected profit order Q
*


For the discrete demand case take the smallest value of Q
*

that satisfies the condition

P(D s Q
*
) > (p - c + g)/(p - s + g)

For the continuous demand case find the Q
*
that solves

F(Q
*
) = (p - c + g) /(p - s + g)

The Optimal Solution
98
THE SENTINEL NEWSPAPER
Management at Sentinel wishes to know how
many newspapers to put in a new vending
machine.
Data
Unit selling price is $0.30
Unit production cost is $0.38.
Advertising revenue is $0.18 per newspaper.
Unsold newspaper can be recycled and net $0.01.
Unsatisfied demand costs $0.10 per newspaper.
Filling a vending machine costs $1.20.
Demand distribution is
discrete uniform between
30 and 49 newspapers.
99
SENTINEL - Solution
Input to the optimal order quantity formula

p = 0.30
c = 0.20 [0.38-0.18]
s = 0.01
g = 0.10
K = 1.20
The probability of the optimal service level =
p+ g - c
p+ g - s
0.30 + 0.10 - 0.20
0.30 + 0.10 - 0.01
= 0.513
=
100
1.0
0.50
0.55
30 49
0.513
39 40
P(D s 39) = 0.50
P(D s 40) = 0.55
Q
*

= 40
SENTINEL Solution
Finding the optimal order quantity Q
*
101
=(B5+B8-B6)/(B5+B8-B7)
=ROUNDUP(B10+E5*(B11-
10),0)
=(E6-B10+1)/(B11-B10+1)
SENTINEL Spreadsheet Solution

102
WENDELLS BAKERY
Management in Wendells wishes to determine
the number of donuts to prepare for sale, on
weekday evenings
Data
Unit cost is $0.15.
Unit selling price is $0.35.
Unsold donuts are donated to charity for a tax credit
of $0.05 per donut.
Customer goodwill cost is $0.25.
Operating costs are $15 per evening.
Demand is normally distributed
with a mean of 120, and a
standard deviation of 20 donuts.
103
WENDELLS BAKERY - Solution
Input to the optimal order quantity formula
p = $0.35
c = $0.15
s = $0.05
g = $0.25
K = $15.00
The optimal service level =
p+ g - c
p+ g - s
0.35+ 0.25 - 0.15
0.35+0.25 - 0.05
= 0.8182
=
104
.8182
=120
Q
*

From the relationship F(Q
*
) = 0.8182 we find the
corresponding z value.
From the standard normal table we have z = 0.3186.
The optimal order quantity is calculated by

Q
*
= + zo


For Wendells Q
*
= 120 + (0.3186)(20) ~ 138
WENDELLS BAKERY - Solution
Finding the optimal order quantity
105
EP(Q
*
) = (p - s) - (c - s)Q
*
- (p + g - s) (o)L[(Q
*
- ) /o| - K

For the normal distribution


L [(Q
*
- ) /o| is obtained from the partial expected value
table.
For Wendells
EP(138) = (0.35 - 0.05)(120) - (0.15 - 0.05)(138) - (0.35 + 0.25
- 0.05)x(20)L[(138 - 120) / 20] - 15 = $6.10
L(0.9) = 0.1004
WENDELLS BAKERY - Solution
Calculating the expected profit
106
=(B5-B7)*B10-(B6-B7)*E6-(B5+B8-B7)*B11*(EXP(-
(((E6-B10)/B11)^2)/2)/((2*PI())^0.5)-((E6-B10)/B11)*(1-
NORMSDIST((E6-B10)/B11)))-B9
=NORMINV(E5,
B10,B11)
=(B5+B8-
B6)/(B5+B8-B7)
WENDELLS BAKERY -
Spreadsheet Solution
107
WENDELLS
The commission strategy
When commission replaces fixed wages
Compare the maximum expected profit of two strategies:
$0.13 commission paid per donut sold,
$15 fixed wage per evening (calculated before).
Calculate first the optimal quantity for the alternative
policy.

Check the expected difference in pay for the operator.
108
WENDELLS
The commission strategy - Solution
The unit selling price changes to
c = 0.35 - 0.13 = $0.22

The optimal order:
F(Q
*
) = (0.22 + 0.25 - 0.15) / (0.22 + 0.25 - 0.05)= 0.7616.

Z = .71
Q
*
= + zo = 120 + (0.71)(20) ~ 134 donuts.
109
Will the bakerys expected profit increase?
EP(134) = (0.22 - 0.05)(20) - (0.15 - 0.05)(134) - (0.22 +
0.25 - 0.05)x(20)L[(134 - 120) / 20] = $5.80 < 6.10

The bakery should not proceed with the alternative
plan.

WENDELLS
The commission strategy - Solution
110
Comments
The operator expected compensation will increase,
but not as much as the bakerys expected loss.
An increase in the mean sales is probable when the
commission compensation plan is implemented. This
may change the analysis results.
WENDELLS
The commission strategy - Solution
111
Copyright 2002 John Wiley & Sons, Inc. All rights
reserved. Reproduction or translation of this work beyond
that named in Section 117 of the United States Copyright
Act without the express written consent of the copyright
owner is unlawful. Requests for further information should
be addressed to the Permissions Department, John Wiley &
Sons, Inc. Adopters of the textbook are granted permission
to make back-up copies for their own use only, to make
copies for distribution to students of the course the
textbook is used in, and to modify this material to best suit
their instructional needs. Under no circumstances can
copies be made for resale. The Publisher assumes no
responsibility for errors, omissions, or damages, caused by
the use of these programs or from the use of the
information contained herein.