Académique Documents
Professionnel Documents
Culture Documents
qt Rr dt Rt
2
1
2
1
.
2
1
0
Area of inventory triangle OAP
Cost of holding inventory during time t= C1 Rt
2
Ordering cost to place an order= C
3
Total cost during time t= C1 Rt
2
+ C
3
Average total cost per unit time, C(t) =
t
C
Rt C
3
1
2
1
+ (12.1)
C will be minimum if
0
) (
dt
t dC
and 0
) (
2
2
dt
t C d
is positive
Differentiating equation (12.1) twice w. r. t. t
3
3
2
2
2 ) (
t
C
dt
t C d
, which is positive for value of t given by the above equation
Thus, C(t) is minimum for optimal time interval.
R C
C
t
1
3
0
2
Optimum quantity q
0
to be ordered during each other
1
3
0 0
2
C
R C
Rt q
Which is known as the optimal lot size ( or economic order quantity) formula due to
R. H. Wilson. It is also called Wilsons or square root formula.
Any other order quantity will result in a higher cost.
The resulting minimum average cost per unit time.
3
1
3
1
3
1 0
2
2
2
1
) (
C
R C
C
R C
C
R C q C +
R C C R C C R C C
3 1 3 1 3 1
2
2
1
2
1
+
(12.4)
Also the total minimum cost per unit time, including the cost of the time.
CR R C C +
3 1
2
(12.4a)
Where, C is cost/ unit of the item.
Equation (12.1) can be written in an alternative form by replacing t by q/R as
q
R C
q C q C
3
1
2
1
) ( +
(12.5)
It may be realized that some of the assumptions made are not satisfied in actual practice.
For instance, it is seldom that a customer demand is known exactly and that
replenishment time is negligible.
Corollary 1: In the above model if the order cost is C3 + bq instead of being fixed,
where b is the order cost per unit item, we can prove that there is no change in the
optimum order quantity due to the changed order cost.
Proof: The average cost per unit time,
) (
2
1
) (
3 1
bq C
q
R
q C q C + +
[From eq.
(12.5)]
For the minimum cost
0
) (
dq
q dC
and
0
) (
2
2
dq
q C d
is positive
i.e.
0
2
1
2
3
1
q
RC
C
or
1
3
2
C
RC
q
3
2 ) (
3
2
2
q
RC
dq
q C d
+ + + +
2 2
D
L R B
RD
LR B p
(12.7)
Assumptions in EOQ Formula
Following simplifying assumptions have been made while deriving the economic
order quantity formula:
1. Demand is known and uniform (constant).
2. Shortages are not permitted; as soon as the stock level becomes zero, it is
instantaneously replenished.
3. Replenishment of stock is instantaneous or replenishment rate is infinite.
4. Lead time is zero. The moment the order is places, the quantity ordered is
received.
5. Inventory carrying cost and ordering cost/order remain constant over time. The
former is linearly related to the quantity ordered and the latter to the number of
orders.
6. Cost of the item remains constant over time. There are no price-breaks or quantity
discounts.
Limitations of EOQ Formula
The EOQ formula has a number of limitations. It has been highly controversial since a
number of objections have been raised regarding its validity. Some of them are
1. In practice the demand is neither known with certainty not it is uniform. If the
fluctuations are mild, the formula can be applicable but for large fluctuations it
loses its validity. Dynamic EOQ models, instead, may have to be applied.
2. The ordering cost is difficult to measure. Also it may not be linearly related to the
number of orders as assumed in the derivation of the model. The inventory
carrying rate is still more difficult to measure and even to define precisely.
3. It is difficult to predict the demand. Present demand may be quite different from
the past history. Hardly any prediction is possible for a new product to be
introduced in the market.
4. The EOQ model assumes instantaneous replenishment of the entire quantity
ordered. In practice, the total quantity may be supplied in parts. EOQ model is not
applicable in such a situation.
5. Lead time may not be zero unless the supplier is next-door and has sufficient stock
of the item, which is rarely so.
6. Price variations, quantity discounts and shortages may further invalidate the use of
the EOQ formula.
However, the flatness of the total cost curve around the minimum is an answer to many
objections. Even if we deviate from EOQ within reasonable limits, there is no substantial
change in cost. For example, if because of inaccuracies and errors, we have selected an
order quantity 20% more (or less) than q0 the increase in total cost will be less than 2%.
Model 1 (b) ( Demand Rate Non-Uniform, Replenishment Rate Infinite)
In this model all assumptions are same as in model 1(a) with the exception that instead of
uniform demand rate R, we are given some total demand D, to be satisfied during some
long time period T. Thus demand rates are different order cycles.
Let q be the fixed quantity ordered each time the order is placed.
Number of orders,
q
D
N
If t
1
is the time interval between orders 1 & 2, t
2
is the time interval between orders 2 & 3
and so on, the total time T will be
= t
1 +
t
2
+ + t
n
This model is illustrated schematically in figure 12.3
Holding costs for time period T will be
1 1 2 1 1
)
2
1
......( )
2
1
( )
2
1
( C qt C qt C qt
n
+
T qC t t t qC
n 1 2 1 1
2
1
) ..... (.
2
1
+
And the ordering cost will be
=C
3
. N
q
D
C .
3
, where C
3 is the ordering cost per order.
Total cost equation for fixed order size q will be
q
D
C T qC q C
3 1
2
1
) ( +
(12.9)
For minimum cost,
[ ] 0 ) ( q C
dq
d
and
[ ] ) (
2
2
q C
dq
d
should be positive.
Differentiating equation (12.9) w.r.t. q,
0
2
1
)] ( [
2 3 1
q
D
C T C q C
dq
d
1
3
1
3
/ . 2 2
C
T D C
T C
D C
q
And
2
3
2
2
2
)] ( [
q
D C
q C
dq
d
1
( 12)
Now the total quantity produced during time t
1
is q and the quantity consumed during the
same period is therefore the remaining quantity available at the end of time t
1
is
I
m
=q-Rt
1
R K
RI
q
m
q
R K
R
I
m
1
]
1
+ 1
or
q
K
R K
I
m
1
]
1
Now holding cost per production run i.e. for time period t
1
. . .
2
1
C t I
m
Or
R q
C
C q
K
R K
q C
/
. .
2
1
) (
3
1
+
,
_
Rt q
q
R C
q C
K
R K
3
1
.
) (
.
2
1
+
(16)
For minimum value of C(q),
0
2
1
)] ( [
2
3
1
q
R C
C
K
R K
q C
dq
d
, which gives
1
3
1
3
1
3
2
. . .
. 2
) (
2
C
R C
R K
K
R K
RK
C
C
C R K
RK C
q
And
3
3
2
2
2
)] ( [
q
R C
q C
dq
d
, which is positive
.
2
1
3
0
, (17)
Model 2 (a) (Demand Rate Uniform, Replenishment Rate Infinite, Shortages
Allowed)
This model is just the extension of model1 (a), allowing shortages. Let
R = number of the items required per unit time i.e. demand rate,
C
1
= cost of holding the item per unit time,
C
2
= shortage cost per item per unit time,
C
3
= ordering cost/order,
q = number of items ordered in one order,
q = Rt,
t = interval between orders,
I
m
= number of items that form inventory at the beginning of time interval t. lead time is
assumed to be zero. Fig. 12.5 shows the variation of inventory with time.
Here the total time period T is divided into n equal time intervals, each of value t. the
time interval t is further divided into two parts t1 and t2.
i. e. t = t1 + t2
Where t1 is the time interval during which items are drawn from inventory ant t2 is the
interval during which the items are not filled. Using the relationship of similar triangles,
Model 2 (b) (Demand Rate Uniform, Replenishment Rate Infinite, Shortages
Allowed, Time Interval Fixed)
Time interval t is fixed which means that inventory is to be replenished after every fixed
time t. All other assumptions of model 2 (a) hold good.
Model 2 ( C ) (Demand Rate Uniform, Production Rate Finite, Shortages Allowed)
This model has the same assumptions as in model 2 (a) except that production rate is
finite. Figure shows the variation of inventory with time.
Referring to figure we find that inventory is zero in the beginning. It increases at constant
rate (K R) for time t1 until it reaches a level Im. There is no replenishment during time
t2, inventory decreases at constant rate R till it becomes zero. Shortage starts piling up at
constant rate R during time t3 until this backlog reaches a level s. lastly, production starts
and backlog is filled at a constant rate K R during time t4 till the backlog becomes zero.
This completes one cycle; the total time taken during this cycle is
This cycle repeats itself over and over again.
Now holding cost during time interval t
Now C is a function of six variables Im, s, t1, t2, t3 and t4 but we can derive relationships
which determine the values of Im, t1, t2, t3 and t4 in terms of only two variables q and s.
An inventory policy is given when we know how much to produce i. e. q and when to
start production, which can be found if s is known.
The manufacturing rate multiplied by the manufacturing time gives the manufactured
quantity.
INVENTORY MODELS WITH PROBABILISTIC DEMAND
The models discussed in previous sections are only artificial since in practical situations,
demand is hardly known precisely. In most situations demand is probabilistic since only
probability distribution of future demand, rather than the exact value of demand itself, is
known. The probability distribution of future demand is usually dete4mined from the date
collected from past experience. In such situations we choose policies that minimize the
expected costs rather than the actual costs. Expected costs are obtained by multiplying the
actual costs for a particular situation with the probability or occurrence of that situation
and then either summing or integrating according as the probability distribution is
discrete or continuous.
Model 3 (a) (Instantaneous Demand, Setup Cost Zero, Stock Levels Discrete and
Lead Time Zero)
Let R = discrete demand rate with probability PR,
Im = discrete stock level for time interval t,
t = constant interval between orders,
C1 = over-stocking cost (over-ordering cost). This is opportunity loss
associated with each unit left unsold
= C + Ch V,
C2 = under-stocking cost (under-ordering cost). This is opportunity loss
due to not meeting the demand.
= S-C-C
h
/2 + C
s,
Where C is the unit cost price, Ch the unit carrying cost, Cs the unit shortage cost, S the
unit selling price and V is the salvage value. If value of any parameter is not given, it is
taken as zero.
Production is assumed to be instantaneous and lead time negligibly small. The problem is
to determine the optimal inventory level Im, where R <= Im or R > Im at the beginning of
each time interval. The variation of inventory with time for these two cases is shown in
fig. 12.7(a) and (b)
When R <= Im, as shown in fig. 12.7(a) there are no shortages; when R > Im, as shown in
fig. 12.7(b) shortages occur.
Then the optimal order quantity I
m0
is determined when value of cumulative probability
distribution exceeds the ratio
2 1
2
C C
C
+
by computing
m m
I R I R
p
C C
C
p
<
+
<
2 1
2
1
Model 3 (b) (Instantaneous Demand, No Setup Cost, Stock Levels Continuous, Lead
Time Zero)
In this model, all conditions are same as in model 3(a) except that the stock levels are
continuous (rather than discrete). Therefore, probability f(R) dR will be used instead of
PR where f(R) is the probability density function of the demand rate R.
Then the optimal order quantity I
m0
is determined when the value of cumulative
probability distribution is equal to by computing
Model 4 (a) (Continuous Demand, Setup Cost Zero, Stock Level Discrete, Lead
Time Zero)
This model is similar to model 3(a) with the difference that demand is continuous rather
than instantaneous i.e. withdrawals from stock are continuous rather than instantaneous.
Also the rate of withdrawals is assumed to be constant.
The reorder time is assumed to be fixed and known; hence setup cost is not included in
calculations. Production is assumed to be instantaneous and lead time negligibly small.
The problem is to determine the optimal order level I
m
where R <= I
m
or R > I
m
, at the
beginning of each time period. The variation of inventory with time for these two cases is
shown in Fig 12.8(a) and (b).
When R <= I
m
as shown in Fig. there are no shortages; when R > I
m
as shown in fig.
shortages occur.
It can be shown that for optimum stock level,
Model 4 (b) (Continuous Demand, Setup Cost Zero, Continuous Stock Levels, Lead
Time Zero)
In this model, all conditions are same as in model 4(a) except that the stock levels are
continuous (rather than discrete). Therefore, probability f(R) dR will be used instead of
P
R
where f(R) is the probability density function of the demand rate R.
Then the optimal order I
m0
is given by
INVENTORY MODELS WITH PRICE BREAKS
In the inventory model discussed so far the production or purchase cost per unit was
assumed tpo be constant. It was not considered during their formulation since it did not
affect the level of inventory. In this section we shall consider a class of inventory
problems in which this cost is variable and depends upon the quantity manufactured or
purchased. This usually happens when discounts are offered for the purchase of large
quantities. These discounts take the form of price breaks. For example, the price breaks
may be given as
Re. 1 per item for purchase of items upto 500,
Re. 0.95 per item for purchase of items upto 1,000,
Re. 0.90 per item for purchase of items 1,001 or more
Clearly, the purchase cost C(q) is a variable and is given by the expression
Such a variable cost must be considered in the inventory model. Further, as this variable
production or purchase cost per unit is more appropriate for purchased parts (because of
quantity discounts), we shall, hereafter, refer only to purchased parts and the problem,
then, is to determine
(i) how often the parts be purchased,
(ii) how many units should be purchased at any one time.
Inventory Models with One Price Break
To understand how the optimal order quantity can be determined in such a case, let us
consider example.
EXAMPLE
An automobile manufacturer purchases 2,400 castings over a period of 360 days.
This requirement is fixed and known. These castings are subject to quantity discounts.
Ordering cost is Rs. 70,000/order and storage cost per day is 0.12% of the unit cost.
Determine the optimal purchase quantity if the supplier has offered the following unit
prices for the castings:
Unit price = Rs. 1,000 for q < 1,000
= Rs. 950 for q >= 1,000
Solution
First we calculate the EOQ if the unit price of castings is Rs 950.
Therefore, the manufacturer should place order for 905 units. This, however, will
not be acceptable to the supplier since he will charge Rs. 950/unit only if the order is
placed for 1,000 or more units in an order. Therefore, this is an infeasible solution. To
avail unit price of Rs. 950, the manufacturer must place an order of at least 1,000 units.
Total cost per day for order quantity of 1,000 units
Since the earlier cost is lower, the optimal order quantity is 1,000 units. The total cost
curve (which is a stepped curve) is shown in Fig 12.9
Cleary, the curve shows a marked drop in the total cost due to price discount at a quantity
of 1,000 units. At this level, the total cost is lower than the total cost corresponding to
882 units.
Inventory Models with Multiple Price Breaks
Sometimes the supplier may offer more than one price breaks. The solution procedure for
such problems is the extension of that used for single price break.
We first determine the EOQ with the lowest cost price. If it is feasible i.e. if the value
falls in the slab of the lot size for this cost price, then this is the optimal order quantity. If
it is infeasible we determine EOQ with the next lowest cost price. If this too is infeasible,
the next lowest cost is tried. The procedure is continued till the feasible EOQ is obtained.
The total cost corresponding to this quantity level is then calculated. Also, the total cost
of order quantities corresponding to cutoff points of the subsequent cost prices is
calculated. The quantity corresponding to the minimum of these total costs is the optimal
order quantity.
EXAMPLE
MULTI-ITEM DETERMINISTIC MODEL
This model considers the inventory system consisting of several items. There are
limitations on production facilities or storage capacity or time or money. This limitation
results in an interaction between the different items so that it is not possible to consider
each item separately. However, simple cases can be handled by using the technique of
Lagrange multipliers.
Consider an inventory consisting of n items. For simplicity let us assume that production
is instantaneous, there is no quantity discount and that no shortages are permitted. Further
let us assume that the demand is known and uniform at a rate of R
i
per unit time for the
ith item. Let C
1i
be the inventory holding cost per unit per unit time and C
3i
be the setup
cost per production run for the ith item.
Total cost of ith item per production run
Limitation on Inventories
If now there is a limitation on inventories that restricts the average number of all types of
stocked items to I
m
the cost C (q
1
, q
2
, ., q
n
) must be minimized subject to the restriction
that
Notice that q
i
0
is dependent on
o
the optimal value of . Also for
o
= 0, q
i
0
gives the
solution of the unconstrained case. The value of
o
can be found by systematic trial and
error. By definition < 0 for the above minimization case. Thus if we try successive
negative values of , its optimum value
o
should result in simultaneous values of q
i
0
which satisfy the given constraint in equality sense. Thus determination of
0
automatically yields q
i
0
.
Limitation on Storage Area
Let A be the maximum storage area available for n items and a
i
be the storage area
required by one unit of ith item. If q
i
is the older quantity for ith item, the storage
requirements constraint becomes