INVENTORY MANAGEMENT AND CONTROL

Background
Inventory is investment in stock, whether of raw materials, work-in-progress or finished
goods. There are costs associated with keeping too much or too little inventories. This is
because some inventory costs are inversely related in the sense that an attempt to reduce
one type will raise another type.
E.g. reduction of shortage cost through increased stock level will increase holding costs.

Inventory problem
How much to order (quantity) and when to order (timing) in order to minimize inventory
costs?

Need for stocks

I. Financial considerations i.e. bulk purchases reduce the following costs
- purchase costs because of discounts
- shortage costs since generally average stock will be high with bulk purchases.
- ordering costs because with bulk purchases, number of orders will be smaller.
II. Stock acts as a hedge (shield) against inflation during inflationary times.
III. Stocks are kept for transaction purposes i.e. to match supply and demand.
IV. Stocks are also kept to guard against uncertainties in demand and lead time.

Inventory Costs
There are broadly 4 types of inventory costs: holding or carrying costs, ordering costs,
shortage costs and purchase cost. These are discussed next.

Holding (Carrying) costs

These are costs incurred because a firm owns or maintains stocks [Link]
i) Opportunity cost of money tied up in stock such as interest foregone.

ii) Storage costs. These include warehouse charges, personnel, equipment etc.

iii) Insurance costs against fire, theft, etc.

iv) Security costs. These include investment in security systems, alarms, electric and
razor wire fencing and hiring of security guards.

v) Perishability costs. This is for perishable products such as edibles, newspapers

and other periodicals, flowers, tax reports etc.

vi) Obsolescence costs. This is due to the product being overtaken by superior
technology. This is quite prevalent in the electronic industry e.g. computers, mobile
phones and TV sets.

Ordering costs
These are the costs of getting the product into the firm’s inventory and they are typically
incurred each time an order is made.

Note: Merchandising firms- ordering costs

Manufacturing firms- setup costs

1
Ordering cycle
Requisitioning

Receipt
Purchase order

Transportation /freight

Purchase made

Examples of ordering costs

i) Purchasing department costs
- Personnel
- Equipment
- Communication costs (telephones and internet )
- Consumables (paper, stationery)
ii) Transportation costs
iii) Insurance on transit
iv) Taxes such as custom duties
v) Clearing and forwarding charges
vi) Handling costs
- Loading and offloading
- Pilferages
- Breakages for fragile items.

Shortage costs
These are incurred as a result of the item not being in stock i.e. an item is missing.

Examples
i) Loss of goodwill may lead to loss of customers and subsequently diminished market
share.
ii) Lost contribution
ii) Cost of backorders
ii) Cost of speeding up orders
ii) Cost of idle resources

Purchase cost
This is what is paid to the supplier or seller by the buyer in exchange of the goods. It is
relevant to optimal inventory policy determination due to the presence of discounts.
In summary,
Inventory Total Cost= purchase cost + holding cost + ordering cost + shortage
cost

The objective of any inventory management system or model is to minimize this total cost

2
Inventory Management Models

Deterministic
Models Stochastic
(Probabilistic)
Models
i)Certainty models; factors are known
and usually constant - Models to cope with uncertainty
and the
factors are usually variable
ii) Simple model - More complex models
iii) Not very realistic - More realistic

THE BASIC EOQ MODEL

Characteristics
-Deal with durable (not perishable) products
-Deal with merchandising firms
- Single product model

Assumptions
i) Demand is constant and known with certainty.
ii) Lead time is constant and it is known with certainty.
iii) There are no shortages; hence no stock- out costs.
iv) All items for a given order arrive in one batch /same time (simultaneously/instantenous
arrivals).
v) Purchase cost is constant i.e. no discounts; hence in the basic EOQ model, purchase cost
is irrelevant
since total purchase cost is the same regardless of quantity ordered in a given order.
vii)Holding cost per unit p.a. is constant. This implies that total holding cost is a linear
function of quantity held.
vii)Ordering cost per order is constant irrespective of quantity ordered. This means that
ordering cost per unit is a declining nonlinear function of quantity ordered.

Suppose ordering cost per order is shs.3,000, fill the following schedule for ordering cost
per unit.

Quantity Ordinary cost per

ordered unit
1 ?
10 ?
100 ?
1000 ?

Ordering cost

3
Thus ordering cost per unit is a declining non-linear function of quantity ordered.

EOQ Model Derivation

There are two methods:
- Graphical approach
- Calculus approach

Graphical Method

Costs
Holding Cost
[Link]

Purchase cost

Ordering cost

EOQ
Quantity (Q)

Observations
 The shape of the total cost function is not influenced by the purchase cost; thus
purchase cost is irrelevant for EOQ determination when there are no discounts.
 Total cost is minimum where holding cost = ordering cost

The variables
Let Q = Order quantity per order (the unknown or the decision variable)
D = Annual demand
Ch = Holding cost per unit p.a.
Cp = Unit purchase cost
i = Holding cost expressed as a % of the unit cost of an item.
Co = Ordering cost per order placed

NB: Ch = Cp x i

I. Ordering cost
O.C = Annual number of orders x Co

D x Co
=
Q

II. Holding cost

4
 H.C. = Average stock in the year x Ch

Or

= Average stock in the year x Cp x i

Receipt & usage profile through time

Quantity
Max stock

Usage/sales Receipts Average stock

Times

Average stock = Maximum stock + Minimum stock

2

Hence H.C =
Or

At min TC,

or

Calculus Approach
TC = P.C + Holding C + Ordering cost

5
TC = DCp +

FOC:

or

SOC:

or

Since D, Q, Co are all positive value hence turning point is minimum

Illustration
Demand for part CD 673 used by Samaki Ltd tends to be constant at annual rate of 4000
units. The cost per unit for this part is Sh.200 and the cost of placing an order is
[Link] Ltd estimates that the annual inventory carrying cost of the part expressed
as a percentage of cost of average stock is 20% (sh. 10 per unit p.a.). Lead-time for this
product is 15 days while the firm works 300 days in a year.

Required:
a) Formulate the optimal inventory policy for part CD 673 i.e.
- Quantity to order (EOQ)
- Frequency of ordering and when to order.
- Re-order level/point.
- Total cost associated with the policy.
b) Suppose it actually turns out that
i) Ordering cost per order = Sh.6000 and
ii) Inventory holding cost percentage i = 15% and yet the policy formulated in
(a) above is implemented for a year, determine the cost of prediction error.

Deterministic EOQ models: Presence of Discounts

Advantages of discounts
Taking discounts results in lowering certain inventory costs such as
- purchase cost
- shortage cost since bulk purchase will generally mean higher stock levels on
average.
- ordering cost since bulk purchases will mean fewer orders.

Disadvantages of discounts

6
 Taking of discounts mean bulk purchases and so, generally, holding costs will increase due
to raised average stock level.

Types of discounts
i) Single discount offer
e.g. unit selling prices is Sh.10 but purchases of 100 units and above will get discount of
3%

ii) Multiple discount offer

This is also called price breaks. The supplier provides a list of price- quantity ranges
e.g.
Quantity Unit selling price
(Shs.)
1 – 100 10.00
101 - 200 9.70
201 - 400 9.50
Above 400 9.00

Principles of inventory policy optimization with discounts

I. Purchase cost is relevant unlike when there are no discounts.
II. In order to minimize inventory costs, purchase the least quantity to just qualify for
discount.
This is because as the size of order increases, holding costs increase much faster
than the savings made from decrease in purchase and ordering costs.

Costs
Tc
Holding costs

Ordering cost

Quantity

Observation
As Q ordering costs 0 so that TC approaches holding cost

Importance
At higher order quantities, holding cost is dominant in the TC of inventory whereas ordering
cost becomes increasingly insignificant. Thus, at higher order quantities, management of
stock is more about management of holding cost.

Single discount offer

Example

7
A Company buys 400 units of a product at a purchase cost Sh.5000 per unit and ordering
cost of Sh.2,000 per order placed. The carrying cost is estimated at 24% of cost of an item
p.a. The company has received a 2% discount offer for purchases of 100 or more units.

Required:
i) Determine the best inventory policy for this item.
ii) Determine the discount level at which the firm will be indifferent between taking and
not taking the discount offer and hence advise on the discount offer.

Multiple discount offers (price breaks)

This is an extension of the single discount offer in the sense that a price-quantity schedule
is availed instead of a single offer.

The solution approach can be broken down into the following steps:
I. Calculate EOQ for each price-quantity range.
II. The EOQ calculated on I will fall in one of 3 categories which will be treated differently
as
Follows:

Below range – Ignore the calculated EOQ but find the TC for the least quantity in the
range.
Within range – Evaluate total cost for the EOQ calculated.
Above range – Ignore this range since there will be another range which will yield
lower TC

Illustration
A company buys 30,000 units of an item per year at an ordering cost of Sh.2500 per order
while
holding cost charges are estimated to be 20% of the cost of average inventory p.a.
The following price-quantity schedule is available from the supplier:

Quantity (units) Unit price (Sh)

1 - 2999 20.00
3000 - 4999 19.00
5000 - 6999 17.00
7000 - 8999 15.50
9000 and above 13.50

Required:
Recommend the optimal inventory policy for this item.

Production and Inventory Management for a Manufacturing Firm:

Economic Batch (EBQ) Model
This is also called the economic lot quantity (ELQ) model and also the gradual
replenishment (as opposed to the instantaneous replenishment) model.
Most production technologies are such that production rates, P are higher than usage rates,
U (demand)
e.g. P= 20 U = 15

8
 Thus it is not necessary to have a continuous or an indefinite production of a given item. It
should
be produced in batches, production stopped for some time (production resources being
used
elsewhere), production started again and so on.

Thus, this problem becomes one of determining the production and inventory policy which
is optimal
in terms of minimizing the total of production and inventory costs in a given period of time.

The decision variables or the unknowns

These are:
i) Quantity to produce in a production run (EBQ/ELQ)
ii) Length of a production run, L
iii) Length of a break between production runs, B
iv) The reorder point, ROP of the user department or the customer e.g. in a production line,
the assembly department would be the user whereas the production department would
be the supplier.
Types of Costs
I. Variable cost of production such as direct material and direct labour costs. Ignore
fixed cost since production-cum-inventory policy adopted will not be influenced by fixed
cost. For a merchandising firm, this is equivalent to purchase cost.
II. Holding cost of raw materials, finished goods and work in progress.
III. Set-up costs are the costs of mobilizing the production resources e.g.
- ordering of raw materials and other components
- assembling the work force
- setting up or realigning the machinery
For a merchandising firm, this is equivalent to ordering cost.
Derivation of EBQ Model
Notes
- All assumption made in deterministic EOQ model apply for EBQ model.
-Additional requirement: Production rates are greater than usage rates (P>U).

Total cost = variable cost of production + Holding cost + Set-up cost

Symbols for the Variables

Let Q = quantity produced in a production run
D = Annual demand
Ch = Inventory holding cost per unit p.a
Cp =Unit variable production cost
= Inventory holding cost.

NB = Ch =Cp i

Co – Set up costs per set up (Equivalent to ordering cost for a merchandising firm)
P – Production rate in units per day
U – Usage
L – Length of a production run
B – Length of a break between production runs

Cost Functions
9
i) Variable production cost = D Cp
ii) Holding cost =

Production- Inventory profile through time

Max stock
Production Usage only
& usage
[Link]/ Average stock
2

0 Run 1 Break 1 Run 2 Break 2 Times

L B L B

Average stock = Max. stock + Min. stock = Max. stock + 0

2 2
Max stock = (P – U) L

But Q = PL so that L =

Hence: Maximum stock = (P – U)Q/P

Average stock =

Therefore Holding cost = Cpi =

iii) Set up cost = Annual number of setups x C o

10
Approaches to Optimization
Graphical Method

Costs Total cost

Holding cost

min. TC

Variable production cost

Setup cost
0 EBQ Q

Notes:
1) Variable cost of production does not affect the optimal point of Q since it is the same
for all values of Q
2) Total cost is minimum where holding cost = Set-up cost

Obtaining the EBQ

H.C = S.C

Calculus Approach

TC =

11
FOC

Since D, Co and Q are all +ve quantities; thus the turning point is minimum.

Comparison of EOQ and EBQ Models

Inventory profiles through time

Receipt Usage/sales EOQ– instantaneous replenishment

Simultaneous
Production Usage only
or receipt & usage EBQ– Gradual replenishment

12
NB: EOQ model is a special case of EBQ model where receipt of items is instantaneous
instead of gradual. Instantaneous receipt is mathematically equivalent to an infinite or
extremely high production rate.
i.e. P approaches

Proof:

If we let , then we have:

EBQ =

But

= which is the EOQ model (proved)

Illustration
Kipsoen Company manufactures part B-2000 on a special lathe for use in a continuous
assembly.
The assemblies that use B-2000 are manufactured at a lower rate. This creates time for odd
jobs to be done on the special lathe when it is not being used for part B-2000. When
parts are being run, deliveries are made to the assembly area; otherwise the
assembly department draws parts from inventory.

The following data is given for part B-2000:

Production rate = 4000 pieces a day

Assembly requirements = 1200 pieces a day
Inventory holding cost = Sh.20 per unit per year
Unit variable production cost = Sh.2000
Set up cost = Sh.110000 per set up
Acquisition lead-time = 10 working days
1 year = 250 working days

Required:
a) Calculate the production department’s economic batch quantity.
b) Determine the length of:
i) A production run.
ii) Break between production runs.

13
c) What is the total cost associated with production/inventory policy formulated in (a)
and (b) above?
d) Determine assembly department’s re-order level.
e) Suppose it turns out that actual setup cost is Sh.510,000 per setup and inventory
holding
cost is Sh.16 per unit per year and yet the policy above is implemented for one year,
determine the cost of prediction error.

INVENTORY MODELS UNDER UNCERTAINTY

Background
An assumption in deterministic or certainty inventory models is that demand and leadtime
are known with certainty and are constant, which is unlikely in practice. If either of these
factors is uncertain, then there will be chances of stockouts and hence stockout costs.

If there is probability of shortages then there may be need to keep extra or additional stock
for such an eventuality. These are known as buffer or safety stock.
While safety stock will reduce shortage costs, it will however increase holding costs as
depicted on the following graph.

Costs
Tc
Holding cost of safety stock
[Link]

Shortage (stockout) cost

S*=? Safety stock level

Problem:
What is the best level of safety stock S* which will minimize the total of holding and
stockout costs?

To solve the problem we have two scenarios:

Scenario I: Shortage cost is known
Scenario II: shortage cost is unknown

Known shortage cost – Tabular approach

Data requirements:
1) Shortages cost per unit, which reflects lost contribution, loss of goodwill, etc.
2) Probability distribution of demand during leadtime period

14
 NB: Concern with shortages is only during leadtime period since this is the only time
a shortage can occur.
3) Current reorder point i.e. reorder point without safety stock. This is usually the
average or expected leadtime demand.
4) Annual number of orders.
5) Holding cost per unit per annum.

Illustration
Mariny Ltd has determined that its reorder point is 50 units when there are no safety
stocks. Its carrying
cost per unit per year is £5 and stockout cost is £40 per unit a customer misses.

The co. has experienced probability distribution for inventory demand during leadtime as
shown below:

Number of units Probability

30 0.20
40 0.20
50 0.30
60 0.25
70 0.05

Required:
If the optimal number of orders p.a. is 6, determine the optimal level of safety stock and
associated reorder point.

Inventory receipt and usage profile under uncertainty

Quantity

Q
3
1

4
Safety
stock
Time
2 Shortage

Explanations
1) Lead time demand is normal or as expected; there is neither excess stock nor use of
safety stock before next order arrives.
2) There is a shortage due to a very high demand during the leadtime period i.e. safety
stock has been exhausted hence a shortage occurs.
3) Lower than normal demand during leadtime period; there is excess stock (other than
safety stock) when the next order arrives.
4) Fairly high demand in the lead time period – some (but not all) safety stock used.
15
Exercise
Umoja manufacturing company has compiled data for the last 100 reorder periods for a
purchased component as follows:

Usage during Number of times this

Lead time Quantity was used
90 7
95 10
100 25
105 50
110 6
115 2
100

The company has found the EOQ to be 250 units with an average daily usage of 5 units.
Leadtime is consistent at 20.4 days. Cost of being out of stock is Sh.300 per component
short and annual carrying cost is Sh.40 per unit. The company works 300 days in a year.

Required:
Determine the optimal level of safety stock for the company and hence the revised reorder
level.

Inventory reorder decisions under uncertainty: unknown shortage cost

Often, it may be not be possible to quantity stockout cost e.g. quantification of such
behavioural concepts as goodwill is difficult.
Sometimes stock out cost may not even apply e.g. what is the stockout cost of a life saving
drug which can be purchased for Sh.1000 if a life is lost due to its absence?
In this case, the firm will want to meet its customers’ needs as much as possible; in any
case not below a certain service level.

Service level
This is the percentage of time that the firm will not be out of stock or the probability of not
being out of stock.

Service level = 1 – probability of a shortage (risk level)

Service level + Risk level = 1

Hence in order to recommend the optimal reorder point (and hence safety stock level), we
need the firm to provides two data items:

i) Service level (or risk level)

ii) Probability distribution of leadtime demand

Illustration
Suppose in the preceding illustration (Umoja Company) stockout cost estimate is not
reliable
and service level desired is 95% i.e. risk of a stock out is 5%, what is the optimal level of
safety-stock
and the reorder point?

16
Exercise
After reviewing previous leadtime demand experience, an analyst feels reasonably
confident that
leadtime demand can be quite adequately represented using a normal distribution that has
a mean
of 60 units and a standard deviation of 8 units.

Required:
i) What reorder point will yield a service level of 95%? ,
ii) What is the level of safety stock?
iii) What service level is associated with safety stock of 10 units?
NB Demand may take other probability patterns such as the uniform distribution.

Inventory policy for perishable items:

Use of Marginal Analysis
Background
How much stock of a perishable product should be stocked each period in order to
maximize long-term profit?
- Examples of perishable are edibles (bread, milk, vegetables), newspapers and other
periodicals, fresh flowers, anniversary cards, etc.
- This problem is also known as the newsboy problem (due to the newspaper seller) since
the approach was developed when solving the newspaper vendor problem. The products
are also called single-period items since they are most useful within a given period of
time i.e. they have a shelf life.
- A characteristic of these items is that periodic demand is uncertain but may follow some
pattern which can be translated into a probability distribution.
- In such a case, an approach known as marginal analysis may be used to recommend the
optimal stock level.

Illustration
A newspaper vendor buys each newspaper copy at Sh. 78 from the publisher and sells it for
Sh 90. If a paper is not sold on the particular day, it can be disposed of through other
channels later at Sh. 11 per copy.

From previous experience, the following data has been gathered:

Number of copies sold Number of days

per day
110 10
120 20
130 40
140 70
150 40
160 1
170 10
Total 200

Required:
Determine the level of daily newspaper stock in order for the vendor to maximize long term
profitability.

17
18