Inventory Models

MBL510: Management Science

1
 Introduction
• Inventory modelling deals with determining
the level of a commodity that a business must
maintain to ensure smooth operations.

• There is need to balance the holding cost and

the penalty cost resulting from inventory
shortage.

2
 Introduction
• The nature of demand is the major determining
the choice of the model to be used.

• Demand can be deterministic or probabilistic.

• An inventory policy answers two questions:

1. How much to order?
2. When to order?

3
 Introduction
• The basis of answering these questions is
the minimization of the following
inventory cost function:

Total inventory cost = (purchasing cost) +

(setup cost) +
(holding cost ) +
(shortage cost)
4
 Introduction
Purchasing Cost
• It is the price per unit of an inventory item.

• At times the item is offered on discount if the

order size exceeds a certain amount.

• This becomes a factor in deciding how much

to order.
5
 Introduction
Setup Cost (Ordering Cost)

• It represents the fixed charge incurred when

an order is placed.

• In production it is the time required to setup a

production run.

6
 Setup/ Ordering Costs
• Increasing the order quantity reduces the
ordering costs associated with a given demand
but it increases the holding costs.
• Reducing the order size increases the ordering
costs as it increases the frequency of ordering
to meet a required demand.
• An inventory cost model balances the two
costs.

7
 Introduction
Holding Cost
• It represents the cost of maintaining inventory
in stock.

• It includes the interest on capital, cost of

storage, maintenance (fumigation) and
holding costs (security & insurance)

8
 Introduction
Shortage Cost

• It is the penalty incurred when we run out of

stock.

• It includes potential loss of income and

customer goodwill.

9
 Introduction
• An inventory system may be based on periodic
review.

• This can be ordering every week or every month.

• Orders are placed at the start of each period.

• A periodic review can occur in a gas station where

new deliveries arrive at the start of the week.
10
 Introduction
• An inventory system may be based on a
continuous review.
• A new order is placed when the inventory
level drops to a certain level called the reorder
point.
• A continuous review occurs in retail stores
where items are replenished only when their
level on the shelf drops to a certain level.

11
 Introduction
Demand pattern may assume one of the
following:
• Deterministic and constant (static) with time
• Deterministic and variable (dynamic) with
time.
• Probabilistic and stationary over time.
• Probabilistic and non-stationary over time.

12
 Deterministic Models
1. Classical Economic Order Quantity (EOQ)
Model

2. EOQ Model with Back Orders Allowed

3. EOQ Model with Quantity Discounts

4. The Continuous Rate EOQ Model

13
 Probabilistic Models
• A Single Period Model with no setup cost
• A single Period Model with no setup cost –
initial stock.
• Continuous distributions – uniform
distribution & exponential distribution
• Discrete distributions – Binomial & Poisson

14
 Assumptions of the Classical EOQ Model

• Demand is deterministic and occurs at a

constant rate.
• Replacement is instantaneous.
• All costs coefficients are constant: unit cost,
holding cost & ordering cost.
• No shortages are allowed.
• No quantity discounts are allowed.

15
 Example 1
Neon lights on the University of Africa campus
are replaced at the rate of 100 units per day.
The physical plant orders the neon lights
periodically. It costs $100 to initiate a
purchase order. A neon light kept in storage is
estimated to cost about $0.02 per day. The
lead time between placing and receiving an
order is 12 days. Determine the optimal
inventory policy for ordering the neon lights.
16
 Example 2
McBurger orders ground meat at the start of each
week to cover the week’s demand of 300kg. The
fixed cost per order is $20. It costs about $0.03
per kg per day to refrigerate and store the meat.
1. Determine the inventory cost per week of the
present ordering policy.
2. Determine the optimal inventory policy that
McBurger should use, assuming zero lead time
between the placement and receipt of an order.

17
Example 3: (Back orders allowed)
Each year, the Smalltown Optometry Clinic
sells 10 000 frames for eyeglasses. The clinic
orders frames from a regional supplier, which
charges $15 per frame. Each order incurs an
ordering cost of $50. Smalltown Optometry
believes that the demand for frames can be
backlogged and that the cost of being short
one frame for one year is $15 (because of loss
of future business).
18
 Example 3: (Back orders allowed)
The annual holding cost for inventory is 30c per
dollar value of inventory.

1. What is the optimal order quantity?

2. What is the maximum shortage that will occur?

3. What is the maximum inventory level that will

occur?
19
 Example 4: Quantity Discount
Dubecar specialises in fast automobile oil change.
The garage buys car oil in bulk at $3 per gallon. A
discount price $2.50 per gallon is available if
Dubecar purchases more than 1000 gallons. The
garage services approximately 150 cars per day and
each oil change takes 1.25 gallons. Dubecar stores
bulk oil at the cost of $0.02 per gallon per day.
Also, the cost of placing an order for bulk oil is $20.
There is a 2-day lead time for delivery. Determine
the optimal inventory policy.
20
 Example 5: (Continuous Rate)
Macho Auto Company needs to produce
10000 car chassis per year. Each is valued at
$2000. The plant has the capacity to produce
25000 chassis per year. It costs $200 to set up
a production run, and the annual holding cost
is 25c per dollar inventory. Determine the
optimal production run size. How many
production runs should be made each year?

21
A Single Period Model with no Setup Cost

• Demand for a period is a random variable

having a known probability distribution.
• Examples of a single period inventory of an
item are:
1. an item becomes obsolete quickly such as a
daily newspaper or a bicycle in the example.

22
 A single Period Model

2. spoils quickly, such as vegetables.

3. is stocked only once e.g. Spare parts for a

single production run.

4. has a future that is uncertain beyond a single

period.
23
 A single Period Model
A trade-off is needed between

(a) The risk of being short and thereby incurring

a shortage cost.

(b) The risk of having excess and thereby

incurring wasted costs or ordering and
holding excess units.
24
 A single Period Model
We consider a general inventory model:
1. Items are purchased (or produced) for a
single period at a cost of “c” dollars per item.
2. The holding cost, the net unit cost of storing
items minus their salvage value is “h” dollars
per item.
3. The cost of unsatisfied demand is given by
“p” dollars per unit (p>c).

25
 A single Period Model
4. There is no initial inventory on hand.

5. Denote by “y” the quantity purchased

(or produced) at the beginning of the period.

6. Let “D” be a random variable that

denotes the demand during the period.

26
 A single Period Model
N.B: One reasonable criterion is to choose the
inventory level that minimises the expected
value of the sums of the two costs:

i. Shortage costs
ii. Holding costs

27
Example 6: Uniform Distribution
A newspaper stand purchases newspapers for
20 cents each and sells them for 25 cents per
newspaper. The dealer buys papers at retail
price to satisfy shortages. The holding cost is
0.1 cent. The demand distribution is a
uniform distribution between 200 and 300.
Find the optimal number of papers to buy.

28
Example 6(b):Exponential distribution

Suppose there is a wholesale distributor of 10

speed bicycles. Suppose that this distributer
offered very favourable terms on the purchase
of a model of a name-brand bicycle whose
production is to be discontinued.
This opportunity appears to be ideal for the
forthcoming season, where, because production
has been discontinued, the stores have been
informed that no reorders are possible.
29
 Example 6(b)
• The cost of each bicycle is $20 and it will be
assumed that there will be no setup cost
incurred.
• The cost of maintaining an inventory is -$9 per
bicycle. This cost includes $1, which represents
the cost of capital tied up, warehouse space,
and so on, and -$10, which is what the
distributor can get for each bicycle remaining in
the inventory after the Christmas season.
30
 Example 6(b)
• Storage cost + salvage value = negative holding
cost.
• Each bicycle is sold for $45 for a profit of $25.
• The penalty cost (unsatisfied demand unit
cost) is considered to be $45 (selling price).
• Lets assume that the demand has an
exponential distribution given by:

31
Example 6(b)

32
 Example 6b)
• Find the optimum number bicycles that the
wholesaler should purchase.

33
 Alternative Formula - Exponential
Distribution
Whenever the demand is exponential with
expectation (lambda), optimal order quantity
can be obtained from the relation:

34
Example 7: (Discrete Distribution)
Chikwasha’s Bakery has the problem of
determining the number of loaves of bread to
stock each day. Chikwasha supplies two types
of bread, white and brown. From past data,
Chikwasha collected the information shown in
Table 1.

35
 Example 7: Table 1
Number of Loaves Sold Sales Frequency (%)
White Brown White Brown
90 60 7 5
91 61 10 8
92 62 12 10
93 63 16 15
94 64 14 15
95 65 10 10
96 66 9 10
97 67 8 8
98 68 6 7
99 69 5 6
100 70 3 6
36
 Example 7

• Chikwasha’s accountant has supplied the

following information in Table 2.

• Find the optimum number of loaves that

should be ordered by Chikwasha.

37
 Example 7: Accountant Data
Bread Penalty Price Holding Cost ($)
($) cost/unit/day
($)
White 300 1.50 250
Brown 350 1.50 260

38
 Single Period Model With Initial Stock
• Suppose that the distributor has 500 bicycles
of the aforementioned type on hand.

• How does this stock influence the optimal

inventory policy?

39
 Model With Initial Stock
• Suppose that the initial stock level is given by
“x”.

• The problem is to determine how much

should be made available , “y”, at the
beginning of the period.

40
 Model With Initial Stock
• Thus, (y – x) is to be ordered

• Amount available (y) = initial stock (x) +

amounted ordered (y -
x)

41
 Model With Initial Stock
The inventory policy which satisfies for p>c, is
given by:

If x<y(0), order up to y(0) (order y(0)-x)

If x>y(0), do not order.

42
 Example 1 - Bicycle Problem
• In the bicycle problem, if there are 500
bicycles on hand, the optimal order policy is to
order up to 11,856 bicycles.

• This implies ordering 11,856 – 500 = 11,356

additional bicycles.

43
 Example 2 - Bicycle Problem
• If on the other hand, there are 12,000
bicycles already on hand (x = 12,000), the
optimal policy is not to order.

44