SCHOOL OF ENGINEERING AND TECHNOLOGY

DEPARTMENT OF AGRICULTURAL ENGINEERING

AE 413 Lecture 4 – Inventory Models

Sectional Outline: Inventory Planning and Management: overview of traditional inventory control
models – deterministic models (EOQ, Production lot-size, Backorder), ABC inventories
classification, periodic review policy, and materials requirements planning; Just-in-Time (JIT)
approach to inventory management.

1. INTRODUCTION

Expenses associated with financing and maintaining inventories are substantial part of the cost of
doing business for most organization/companies. In many companies, especially those with
expensive products, the cost associated with raw materials, in process, and finished goods
inventories can run into millions of shillings.

There are a number of costs associated with maintaining or carrying a given level of
inventory/stocks. These costs taken together are usually referred to as the inventory holding costs.
Inventory holding costs include: costs of financing or cost of capital (if borrowed funds = interest
charge; if firm’s own money = opportunity costs), insurance, taxes, breakages/pilferage, and
warehouse overheads which depend for most part on the value of the inventory.

We can, therefore, define inventory as any idle goods or materials that are waiting to be used. The
managers are faced with dual problems – kind of dilemma - of maintaining sufficient inventories to
meet the demand for goods and at the same time incurring the lowest possible inventory costs.

Basically, managers attempt to solve these problems by making the best possible decisions with
respect to the following: (a) how much should be ordered when the inventory for a given item is
replenished? (b) when should the inventory for a given item be replenished? It is a question of how
much to order and when to order.

The answers to these questions, over the years, have been provided through the use of what we will
be referring to as inventory control models. We will look at traditional/classical inventory control
models – the deterministic models (Economic Order Quantity; production-lot-size; and Backorder
model), ABC inventories classification, periodic review policy and materials requirement
planning, and the more recent approaches to inventory control (Just-in-Time) management.

2. THE INVENTORY MODELS

2.1 Economic Order Quantity (EOQ) Model

This is the best known and most fundamental inventory decision model. The model is potentially
applicable when the entire quantity ordered arrives in the inventory at one point in time and when
the demand for the item has a constant, or nearly constant rate (for example, 5 units per day, 500
units per week).

1
The inventory holding costs, the order costs and the demand information are the three data items
that need to be known prior to the use of the EOQ model.

Let us develop the total cost model:

 we start by defining the symbol Q to be the size of the order quantity.
 therefore, the how much to order decision, involves trying to find that value of Q which
will minimize the sum of inventory holding and ordering costs.
 The general inventory pattern during several order cycle periods of length T can be
represented as shown in Figure 1.
 If the average inventory during each cycle is ½ Q, the average inventory level over any
number of cycles is also ½ Q.
 The inventory holding costs can be calculated using the average level. Thus we can
calculate the inventory holding cost by multiplying the average inventory by the cost of
carrying one unit in inventory for a stated period (normally one year – i.e. on annual
basis).
 If we let: I = annual inventory carrying charge; C = unit cost of inventory item (purchase
price), then: the cost of storing/ holding one unit in inventory for a year, denoted by, Ch is
given by: Ch = IC.
 Therefore, the general equation for the annual inventory holding cost is:
Annual inventory holding cost = Average inventory x annual holding cost per
unit = ½ QCh.

 To complete our total cost model, we must now include the ordering cost. The goal is to
express this cost in terms of the order quantity Q. And since the holding cost was expressed
as an annual cost, we need to express ordering costs on annual basis as well.
 The first question will be, how many orders will be placed during the year: If we let: D
denote demand for the item/product for the year and if we place order for Q units every
time we place an order, then
we will have to place (D/Q) orders during the year.

2
 Figure 1: Graphical Representation of the EOQ Model.

 If Co is the cost of placing one order; the general equation for the annual ordering
cost is:
Annual Ordering Cost = No. of Orders per Year x Cost per
Order = (D/Q) x Co.

 Thus:
The total annual cost (TC), i.e., inventory holding cost plus the
ordering cost, can be expressed as:
TC = ½ Q Ch + (D/Q) Co.

We have with the above equation; been able to express the total annual cost as a function of
one of the decisions, how much should we order. The next step is to find the order quantity
Q that does minimize the total cost as stated in the above total cost equation. We can find
the order quantity Q that minimizes the total cost by setting the first derivative dTC/dQ
equal to zero and solving for Q*:

dTC/dQ = ½Ch – (D/Q2) Co = 0;

½Ch = (D/Q2) Co;

ChQ2 = 2DCo;

3
 Q2 ( 2DCo)/Ch

Hence: The Economic Order Quantity Formula:

Q* = √ ¿ ¿)]
Note: The second derivative, i.e. d2TC/dQ2 = (2D/Q3) Co is greater than zero, and therefore
the Q* equation above is in fact the minimum cost solution.

 The when to buy decision can now be answered as we already know the how much
to order. We express the when to order decision in terms of the re-order point. The
Re-order Point is the inventory level at which a new order should be placed.
 When an order is guaranteed to be delivered for a specified time period from the
time an order is placed, then, this time is referred to as the Lead Time for a new
order.
 The anticipated demand during the lead-time period is referred to as the Lead Time
Demand.
 For inventory systems using a constant demand rate assumption and a fixed lead-
time, the reorder point is the same as the lead-time demand.
 The general expression for the re-order point is:
Re-order Point = Demand per day x lead-time in days.
 The question of how frequently the order should be placed can now be answered.
This period between orders is referred to as Cycle Time. The general expression for
the cycle time of T days is given by:

T = 365/(D/Q*) = (365Q*)/D.
Note: the number of orders that will be placed in a year is D/Q*.

Inventory Control Costs Tradeoff

The graphical depiction of the inventory costs, i.e. holding, ordering and total costs is as shown in
Figure 2.

4
 Figure 2: Inventory Costs Relationships

Sensitivity Analysis

 Note that parameters, such as cost per order, and inventory holding costs, etc. are just good
estimates. There is a need to consider how much the order size recommendation would
change if estimated ordering cost and holding costs vary.

2.2 Production Lot Size Model

This is similar to the Economic Order Quantity Model where we are attempting to determine how
much to order or produce and when the order should be placed. The assumption of constant
demand rate also applies; however, instead of the inventories arriving at the store/warehouse in
quantity Q*, as assumed in the EOQ model, we assume the inventory is received at a constant rate
over a period of time.

This kind of model applies to production situation where once an order is placed, production
begins and a constant number of units of an item is added to the inventory per period, say each day
of production until production run is completed. This model is also referred to as the production
lot-size or lot-quantity or batch quantity or economic production quantity model, whose pattern is
as shown in Figure 3.

5
 Q

Production Usage Production Usage Production

and usage only and usage only and usage
Q*
Cumulative
production
Imax

Amount
on hand

Time

Figure 3: Production Lot Size Model

Let us build the production lot-size model:

We let:d = daily demand rate for the product,

p = daily production rate for the item/product,
t = number of days for a production run

Since we are assuming ’p’ will be larger than ‘d’, then ‘p−d’ will be the daily rate
of inventory build-up and if we run production for ‘t’ days, and place ‘p−d’ in
inventory every day, the inventory level at the end of the production run will be
(p−d) × t which will be the maximum inventory.

If we are producing quantity Q units of the product at a daily rate of ‘p’ units,
then: Q = p × t, and we can compute the length of the production run ‘t’ to be: t = Q
÷ p days.
Thus:
Max Inventory = (p – d) × t = (p−d) × (Q ÷ p) = [(1 – (d ÷ p)] × Q.

Therefore: Avg. Inventory = ½ [(1 – (d ÷ p)] × Q.

And where: Annual Inventory holding cost, Ch = IC per unit, as in EOQ model, then:

Ann. Inventory Holding Cost = (Average Inventory) × (Annual Holding Cost per unit) =
½ [(1 – (d ÷ p)] × Q × Ch

If D is the annual demand for the product and Co is the set-up cost per production run, then:

Total Ann. Set-up Cost = (No. Production Runs Per Year) × (Set-up Cost per Run) = (D
÷ Q) × Co

6
 Therefore: The Total Annual Cost, TC is given as:

TC = ½ [(1 – (d ÷ p)] × Q × Ch + (D ÷ Q) × Co
If we operate 270 days in a year for production, then daily demand, ‘d’ in terms of annual
demand, ‘D’ could be written as: d = D ÷ 270,

If we let, P denote the annual production, if it were produced everyday, then:

P = 270× p or p = P ÷ 270; and therefore: d ÷ p = D÷ P; and

Therefore:
TC = ½ [(1 – (D ÷ P)] × Q × Ch + (D ÷ Q) × Co; and using calculus,

Q* = √ {[(2DCo)] ÷ [(1 – (D ÷ P) × Ch]}

-------------This is the production lot-size or batch quantity model.

2.3 Inventory Control Model with Planned Shortages

In many inventory situations, a shortage or stockout – a demand that cannot be met or supplied
from inventory or production – is undesirable and should be avoided if possible. In order to avoid
stockout, many organizations normally resort to putting in place a safety stock policy. The policy
arrangements are depicted diagrammatically as follows which is self-explanatory:

However, there are other cases in which it may be desirable, from an economic point of view, to
plan for and allow shortages. In practice this type of situation is normally found where the value
per unit of the inventory is very high and therefore the inventory holding cost is prohibitively high.
An example of this kind of situation is high value agricultural machinery dealership. For example,
a tractor dealer will not hold in stock fifty tractors expecting people to come and buy! The specific
model that we are going to look at that allows for planned shortages – is known as a ‘backorder’
model.

Back-order Model

In backorder situation an assumption is made that when a customer places an order only to find out
that the order cannot be fulfilled (the item is out of stock), the customer does not withdraw the
order rather waits until the next shipment arrives for his order to be filled.

Let us now develop the model as an extension of the EOQ model. The assumptions of goods
arriving in inventory all at one time, and a constant demand applies:

7
 If we let S indicate the amount of shortage or number of backorders accumulated when a
new shipment of size Q is received, then the inventory system has the following
characteristics:
o With S backorders existing when Q is received, the S backorders are immediately
filled and the remaining Q−S units are placed in inventory.
o The Q−S will be the maximum inventory level for the system.
o The inventory cycle T will, therefore be divided into two distinct time periods; t 1
days or weeks when the inventory is on hand and orders are filled as they are
received and t2 days or weeks or months when there is stockout and all orders
received are placed on backorder.

The inventory pattern for this model, where negative inventory represents the number of
backorders is as shown in Figure 4.

With the inventory pattern defined, let us now build the total cost model. For the inventory
model with backorders, in addition to the usual inventory holding costs and
ordering costs, we will also incur backordering costs in terms of labor and special delivery costs
directly associated with backorders. It is customary to express all backorders costs in terms of how
much it costs to have a unit on backorders for a stated period of time.

Therefore:
o Let us use a hypothetical example to suggest a procedure of how to calculate
average inventory level. If we have an average inventory of 4 units for three days
and no inventory on the fourth day, what is our average inventory level over the
four-day period?

Inv.
Level

S
Time

8
 Figure 4; Planned shortages model –Backorder Model

We should say…

4 units ( 3 days ) +0 units (1 day) 12

= =3units
4 days 4

Similarly, with a maximum inventory of Q−S units during the t1 days we have
inventory, an average inventory of (Q−S) ÷ 2 is on hand for the t1 days. No
inventory will be carried for the t2 days in which we experience stockout
(backorders).

So, over the total cycle time of T = t1 + t2 days, we can calculate the average
inventory level as follows…

Avg. inventory level = ½ (Q−S)t 1+ 0 t 2 ¿/T =(½(Q−S)t 1)/T

But if the maximum inventory is (Q−S) and we let d represent the constant daily
demand, we have…
t1 = (Q−S)/d days … (Max. inventory level will be used up in t1 days)

Since Q units are ordered each cycle, we know the length of the cycle will be…
T = Q ÷ d days
Therefore, …
Avg. inv. Level = ½ (Q−S)[(Q – S) ⁄ d ]/(Q ÷ d) = (Q – S)2 /2Q
Thus the average inventory level is expressed in terms of the two inventory
decisions, how much to order (Q) and the maximum number of backorders we will
allow (S).

If we let d represent the annual demand, we have…

Annual number of orders = D⁄Q

The next step is to develop an expression for the average backorder level, similarly…
We have an average number of backorders during the period t2 of ½ of the
maximum number of backorders or ½S and since we have no backorders
during the period t1 days when we have inventory in stock, we can calculate
average backorder level in a similar manner…
Avg. backorder level = (0 t 1+(S ⁄ 2)t2 )/T =(S ⁄ 2)t2 /T
Since we let the maximum number of backorders reach an amount S at a rate of d,
the length of the backorders portion of the inventory cycle is..
t2 = S⁄d. Hence, by substitution…

9
 Average backorder level = ((S ⁄ 2)(S ⁄ d )) /(Q ⁄ d )=S2 ⁄ 2 Q
If we let Cb = cost to maintain one unit on backorder for one year, then our total
annual cost equation (TC) for the inventory model with backorders becomes:

( Q – S )2 D S2 C
TC = 2Q
C h+ C o +
Q 2Q
b

Given the cost Ch, Co, and Cb, the annual demand D, we can, using calculus,
determine the minimum cost values for our inventory decisions, Q and S as follows:

Optimal Order Quantity:

Q* =
√(
[
2 D Co
Ch )( C h +C b
Cb )]

And
Ch
Planned shortages: S* = Q* ( Ch +¿+C ¿ .
b

Note:
1. As the backordering cost Cb , becomes large relative to the inventory holding cost Ch , the
quantity Ch ⁄(Ch + Cb ) in the equation for planned shortages, will be relatively small. In this
case the backorder model and the EOQ model provide very similar results. 2. As the holding
costs Ch (Ch = IC), become large, the number of backorders S becomes larger. This explains
why items with very high value (high per unit cost, C) are handled on a backorder basis.

3. THE ABC ANALYSIS

Is a business term used to define an inventory categorization technique often used in materials
management. It is also known as Selective Inventory Control.
The policies based on ABC analysis:
• A Items: very tight control and accurate record keeping is required;
• B Items: less tightly controlled and reasonably good records required; and
• C Items: simplest controls possible and minimal records required to be kept.

The ABC analysis provides a mechanism for identifying items that will have a significant impact
on overall inventory cost, while also providing a mechanism for identifying different categories of
stock that will require different management and controls. The ABC analysis (Figure 5) suggests
that inventories of an organization are not of equal value. Thus, the inventory is grouped into
three categories (A, B, and C) in order of their estimated importance.
'A' items are very important for an organization. Because of the high value of these
‘A’ items, frequent value analysis is required. In addition to that, an organization needs to
choose an appropriate order pattern (e.g., ‘Just- in- time’) to avoid excess capacity. 'B' items are
important, but of course less important than ‘A’ items and more important than ‘C’ items.

10
Therefore ‘B’ items are intergroup items. 'C' items are marginally important.
There are no fixed threshold for each class, different proportion can be applied based on objective
and criteria. ABC Analysis is similar to the Pareto principle in that the 'A' items will typically
account for a large proportion of the overall value but a small percentage of number of items.

An example of an ABC class is:

‘A’ item – 20% of the items accounts for 70% of the annual consumption value of the
items.
‘B’ items - 30% of the item’s accounts for 25% of the annual consumption value of the
items.
‘C’ items - 50% of the item’s accounts for 5% of the annual consumption value of the
items.
Another recommended breakdown of ABC class is:
"A" approximately 10% of items or 66.6% of value
"B" approximately 20% of items or 23.3% of value
"C" approximately 70% of items or 10.1% of value.

Prioritization of the management attention (Figure 6)

Inventory optimization is critical in order to keep costs under control within the supply chain. Yet,
in order to get the most from management efforts, it is efficient to focus on items that cost most to
the business. The annual consumption value is calculated with the formula:
(Annual demand) x (item cost per unit). Through this categorization, the supply
manager can identify inventory hot spots, and separate them from the rest of the
items, especially those that are numerous but not that profitable.

The following steps explain the classification of items into A, B and C categories: 1. Find out
the unit cost and and the usage of each material over a given period. 2. Multiply the unit cost
by the estimated annual usage to obtain the net value. 3. List out all the items and arrange them
in the descending value. (Annual Value). 4. Accumulate value and add up number of items and
calculate percentage on total inventory in value and in number. 5. Draw a curve of percentage
items and percentage value. 6. Mark off from the curve the rational limits of A, B and C
categories.

11
 Figure 5: Graphical Example of ABC categorization

Figure 6: Prioritization of the management attention

4. PROCUREMENT AND WAREHOUSE APPLICATIONS

12
The results of an ABC Analysis extend into a number of other inventory control and management
processes, including:
1. Review of stocking levels – As with investments, past results are no guarantee of
future performance. However, “A” items will generally have greater impact on projected
investment and purchasing spend, and therefore should be managed more aggressively in
terms of minimum and maximum inventory levels. Obsolescence review - by definition,
inactive items will fall to the bottom of the prioritized list. Therefore, the bottom of the “C”
category is the best place to start when performing a periodic obsolescence review.
2. Cycle counting – The higher the usage, the more activity an item is likely to have,
hence the greater likelihood that transaction issues will result in inventory errors. Therefore,
higher priority items are cycle counted more frequently. Generally, “A” items are counted
once every quarter; “B” items once every 6 months; and “C” items once every 12 months.
3. Identifying items for potential consignment or vendor stocking – Since “A” items
tend to have a greater impact on investment, these would be the best candidates to
investigate the potential for alternative stocking arrangements that would reduce investment
liability and associated carrying costs.
4. Turnover ratios and associated inventory goals – By definition, “A” items will have
greater usage than “B” or “C” items, and as a result should have greater turnover ratios. When
establishing investment and turnover metrics, inventory data can be segregated by ABC
classification, with different targets for each category.

Periodic Reviews
To make the most effective use of ABC classifications, the analysis should be
completed at least on an annual basis, and more often as necessary.

5. OTHER INVENTORY CLASSIFICATION TECHNIQUES

HML Classifications

The High, medium and Low (HML) classification follows the same procedure as is adopted in
ABC classification. Only difference is that in HML, the classification unit value is the criterion
and not the annual consumption value. The items of inventory should be listed in the
descending order of unit value and it is up to the management to fix limits for three categories.
For examples, the management may decide that all units with unit value of Tshs 500,000 and
above will be H items, M items are those that are less than Tshs 500,000 but more than
Tshs 200,000 and anything below Tshs 200,000 are L items. The HML analysis is useful
for keeping control over consumption at departmental levels, for deciding the frequency of
physical verification, and for controlling purchases.

VED Classification

While in ABC, classification inventories are classified on the basis of their consumption value
and in HML analysis the unit value is the basis, criticality of inventories is the basis for vital,
essential and desirable categorization.
The VED analysis is done to determine the criticality of an item and its effect on

13
 production and other services. It is specially used for classification of spare parts.
If a part is vital, it is given V classification, if it is essential, then it is given E
classification and if it is not so essential, the part is given D classification. For V
items, a large stock of inventory is generally maintained, while for D items,
minimum stock is enough.

SDE Classification

The SDE analysis is based upon the availability of items and is very useful in the context of
scarcity of supply. In this analysis, S refers to scarce items, generally imported, and those,
which are in short supply. D refers to difficult items that are available indigenously but are
difficult items to procure. Items, that has to come from distant places or for which reliable
suppliers are difficult to come by fall into D category. E, refers to items which are easy to
acquire and which are available in the local markets. The SDE classification, based on
problems faced in procurement, is vital to the lead time analysis and in deciding on
purchasing strategies.

FSN Analysis

FSN stands for fast moving, slow moving and non-moving. Here, classification is based on the
pattern of issues from stores and is useful in controlling obsolescence. To carry out an FSN
analysis, the date of receipt or the last date of issue, whichever is later, is taken to determine
the number of months, which have lapsed since the last transaction. The items are usually
grouped in periods of 12 months. FSN analysis is helpful in identifying active items, which
need to be reviewed regularly, and surplus items, which have to be examined further. Non-
moving items may be examined further and their disposal can be considered.

6. THE JUST-IN-TIME (JIT) APPROACH TO INVENTORY CONTROL IN

PRODUCTION

6.1 Introduction

The whole essence of JIT is to reduce inventory, minimize waste and responding to customer
demands. The primary motivation behind JIT is that inventory represents inefficiency.
Answers to some pertinent questions underlines the JIT concept:
 When is the best time to have an inventory part ready for production? Just in time.
 When is the best time to have an item ready for the next step in production? Just in time.
 When is the best time to have an item ready for delivery to a customer? Just in time.
 So why do manufacturers and retailers build inventory of finished goods and raw
materials? Just in case!

A buffer /safety stock inventory on hand is comforting but costly. If you hold a lot of items in
inventory, you are tying up a huge amount of capital unnecessarily. These items can be lost,

14
 stolen, or damaged or they can deteriorate, they occupy space, which could otherwise be
devoted to other use. Items can become obsolete, particularly when products are improved and
changed often.

In the 1970s, the Toyota company developed a guiding philosophy for manufacturing that
minimized waste and improve quality. The JIT philosophy advocates a lean approach to
production. It consists of producing products or obtaining products from the suppliers at the
instant they are required. Since JIT strives to reduce inventory levels, many people refer to it as
stockless production or zero inventories.
The key benefits of JIT therefore are: low inventory levels, low wastage, high quality
production and high customer responsiveness.

6.2 The JIT systems

To facilitate the JIT approach, you need a variety of systems to be in place, notably the Kanban
and the continuous improvement systems such as the Total Quality Management. Probably the best
method used to implement JIT is the Japanese Toyota’s Kanban system. Kanban a Japanese word
literally meaning signboard or billboard or card. The term kanban combines the Japanese words
“kan”, meaning “visual” and “ban”, meaning “card” or “board”. A kanban is a visual card or other
cue that signals something is needed. It is a “pull” system as opposed to “push” system, where the
manufacturer or user determines the supply.
Notes: Consider a manufacturing process in which a product must pass through four
work stations before the product is completed, beginning at station 1 and ending at
station 4. If we think as if the product is flowing down a river, then station 1 is farthest
upstream, and station 4 is the farthest downstream. In a “push” system, materials push
their way through the system from station 1 to station 4. Thus, if 50 units of the product
are completed at workstation 1, they are then pushed toward workstation 2; causing
work for the employees at that station. Once a unit of the product finishes at workstation
2, it is pushed on to workstation 3, and so on. In contrast, in a “pull” system, an
upstream station does not produce anything unless production is authorized by the
immediately succeeding downstream workstation. For example, workstation 2 cannot
work on any material unless authorized by workstation 3.

One of the
most common applications of Kanban is a bin system with cards. You have one bin on the
plant floor that contain the manufacturing supplies, and one bin waiting in the supplies
inventory. Each bin has a card with production details. When the bin in the plant is empty, the
bin and the Kanban card are taken to supplies inventory. The bin and the Kanban card waiting
in the supplies are taken to the floor to replace the empty bin. The empty bin is sent to the
supplier, who fills it and sends it back to supplies inventory.

Depending on the complexity of the manufacturing process and kanban system, you can have
many different cards in use. You can send a bin and a card from a machine to a machine and
from department to department, depending on how many control points you want to maintain.

15
HOD/AE 413/EngOpnMangt/Jan2024

16