© Majid [email protected].

2020
©inventory control_Lumpy model

DYNAMIC EOQ- PART 1 (LUMPY MODEL)

(DR. MAJID KHAN MAJAHAR ALI)

When to Use EOQ Models

Demand is often irregular, or “lumpy”. If demand is irregular, the
Constant Demand Assumption that was required for all EOQ
models will not be satisfied.

To decide whether demand is sufficiently regular to justify use of

EOQ models, Peterson and Silver (1998) recommend that the
following computations be done:

1. Determine the estimate 𝑑̅ of the average demand per period

given by

1 𝑖=𝑛
𝑑̅ = ∑ 𝑑𝑖
𝑛 𝑖=1

2. Determine an estimate of the variance of the per-period

demand D from
1
Est. var D = ∑𝑖=𝑛 2
𝑖=1 𝑑𝑖 − 𝑑
̅2
𝑛

3. Determine an estimate of the relative variability of demand

(called the variability coefficient). This quantity is labelled 𝑉𝐶,
where

est. var D
𝑉𝐶 =
𝑑̅ 2

If 𝑉𝐶 is small, this indicates that the Constant Demand Assumption

is reasonable.
 © Majid [email protected]
©inventory control_Lumpy model

Research indicates that the EOQ should be used if 𝑽𝑪 < 𝟎. 𝟐𝟎;

otherwise, demand is too irregular to justify the use of an EOQ
model.

Lumpy Demand Models

Demand for many items is not constant or uniform but tends to be
lumpy or non-uniform. This may be due to intermittent
manufacturing or assembly requirements. A number of heuristic
decision rules have been proposed.

1. Lot-for lot
This simplest procedure requires that an order be placed so
that it will be received in any period for which non-zero
demand exist. The size of the order should just equal that
period’s demand or requirement.

2. EOQ
The standard EOQ formula can be used if modified slightly
to fit the lumpy demand situation. A standard order
quantity, 𝑞, is determined from:

2𝐾𝐷
𝑞=√
ℎ𝑝

where
𝐾 = ordering cost
ℎ𝑝 = holding cost/unit/period

𝐷 = average demand per period

 © Majid [email protected]
©inventory control_Lumpy model

An order for this quantity should be placed so as to arrive in

any period for which a negative inventory balance is
otherwise projected. If the projected negative balance is
greater than the standard order quantity, the actual order size
should be increased (which is often done by using multiples
of the EOQ quantity) so that the projected inventory balance
when the order is received is not negative.

3. Periodic Order Quantity (POQ)

This method requires the conversion of the EOQ into time
periods rather than units. If 𝑞 is the EOQ and 𝐷 is the
average demand per time period, then
𝑞
POQ =
𝐷
where POQ would be rounded to the nearest whole number
and represents the number of periods of demands to be
covered by an order. Whenever the projected on-hand
balance is negative, an order should be placed to arrive in
that period. The order size should be sufficient to cover the
number of periods specified by the calculated POQ.
4. Fixed Order Quantity
Frequently, the size of the order may be restricted due to
minimum order requirements (to receive quantity discount),
standard batch size (to fit material transport or process
requirements), etc. Under these rules, an order for that
restricted or fixed quantity is placed so that it will arrive in
any period in which the projected on-hand inventory balance
will be negative otherwise. Multiples of this fixed quantity
 © Majid [email protected]
©inventory control_Lumpy model

is used when the projected negative balance exceeds the

standard order size.
5. Part Period Balancing
This procedure attempts to balance ordering and holding
costs in a manner similar to EOQ. An order is placed so that
it will arrive in any period in which the projected on-hand
balance will be negative otherwise. The order quantity will
cover at least the requirement for that period. The number of
additional periods included in the order (if any) is chosen so
that the carrying costs (for these additional periods) most
closely equal the ordering cost. Do remember the most
economic lot size need to be used. We can calculate the
economic part per period (EPP) using the formulae below;
𝑆𝑒𝑡𝑢𝑝 𝐶𝑜𝑠𝑡
EPP =
𝐻𝑜𝑙𝑑𝑖𝑛𝑔 𝑐𝑜𝑠𝑡
6. Least Unit Cost
This method determines order quantities in a way that
minimizes the per-unit ordering and holding costs of each
order. As before, an order is placed to arrive in any period
for which the projected on-hand inventory balance is
negative. The number of periods included in the order is
chosen so that the average total ordering and carrying cost
per unit included in the order is minimized. Do remember
the most economic lot size need to be used. We can calculate
the economic part per period (EPP) using the formulae
below;
𝑆𝑒𝑡𝑢𝑝 𝐶𝑜𝑠𝑡
EPP =
𝐻𝑜𝑙𝑑𝑖𝑛𝑔 𝑐𝑜𝑠𝑡
 © Majid [email protected]
©inventory control_Lumpy model

Problem:

The Broadway Electronics Company uses a standardized tape drive unit in

several items that it manufactures. The demand for the unit is lumpy because of
intermittent assembly schedules for various items that use this drive unit. The
company wishes to determine which of several lot sizing rules should be used
for this item. Demand for the coming 12-week period is shown below, based on
the planned end item assembly schedule:

Week
1 2 3 4 5 6 7 8 9 10 11 12
Demand 40 10 20 50 40 0 10 80 15 5 30 25

If ordering costs are $40 per order and holding costs are $0.50 per unit per week,
determine the ordering schedule and total relevant costs for the 12-week period
using the following lot-size policy (assume zero initial inventory):

(a) Lot-for-lot.
(b) EOQ.
(c) Periodic order quantity.
(d) Part period balancing.
(e) Least unit cost.

Answer :
ordering costs, K = $40,
holding costs = $0.50,
Initial Inventory = 0

(a) Lot-for-lot.
(*The size of the order should just equal that period’s demand or
requirement, in here we maximize the setup cost)
Week
1 2 3 4 5 6 7 8 9 10 11 12
Demand 40 10 20 50 40 0 10 80 15 5 30 25
Order 40 10 20 50 40 0 10 80 15 5 30 25
Balance 0 0 0 0 0 0 0 0 0 0 0 0
Total relevant costs
 © Majid [email protected]
©inventory control_Lumpy model

(for 11 order and 0 inventory 9 (from the above table))

= $40 x 11
= $440.00

(b) EOQ (Unknown Demand)

(*hint: find the average demand, next find the EOQ and finally substitute
into the table and find Balance (difference between Demand and
order))
Step 1:

Average demand, D from the table,

𝟒𝟎+𝟏𝟎+𝟐𝟎+𝟓𝟎+𝟒𝟎+𝟎+𝟏𝟎+𝟖𝟎+𝟏𝟓+𝟓+𝟑𝟎+𝟐𝟓
D= = 𝟐𝟕. 𝟎𝟖
𝟏𝟐

Step 2:

2(40)(27.08)
𝑬𝑶𝑸 = √ = 65.82~𝟔𝟔
0.5

Step 3:

Week Total
1 2 3 4 5 6 7 8 9 10 11 12
Demand 40 10 20 50 40 0 10 80 15 5 30 25
Order 66 66 66 66 66 5
Balance 26 16 62 12 38 38 28 14 65 60 30 5 394
Step 4:

Total relevant costs

(for 5 orders and 394 total inventory (from the above
table))
= $40 x 5 + 0.50*(394) = $397.00
 © Majid [email protected]
©inventory control_Lumpy model

(c) Periodic order quantity. (POQ)

(*hint: conversion of the EOQ (using average demand) into time periods
rather than units. If q is the EOQ and D is the average demand per time
period, then, POQ= q/D
where POQ would be rounded to the nearest whole number and
represents the number of periods of demands to be covered by an order.

Step 1:

Average demand, D from the table,

𝟒𝟎+𝟏𝟎+𝟐𝟎+𝟓𝟎+𝟒𝟎+𝟎+𝟏𝟎+𝟖𝟎+𝟏𝟓+𝟓+𝟑𝟎+𝟐𝟓
D= = 𝟐𝟕. 𝟎𝟖
𝟏𝟐

2(40)(27.08)
𝑬𝑶𝑸 = √ = 65.82
0.5

Step 2:

65.82
POQ = = 2.43~2
27.08

Step 3:

Week Total
1 2 3 4 5 6 7 8 9 10 11 12
Demand 40 10 20 50 40 0 10 80 15 5 30 25
Order 50 70 40 90 20 55 6
Balance 10 0 50 0 0 0 80 0 5 0 25 0 170
Step 4:

Total relevant costs

(for 6 orders and 170 total inventory (from the above
table))
 © Majid [email protected]
©inventory control_Lumpy model

= $40 x 6 + 0.50*(170) = $325.00

(d) Part period balancing.

(*hint: (i) Find EPP = (Setup Cost)/(Holding cost)

(ii) Choose so that the holding costs ((for these additional periods)
from the demand table)) most closely equal the ordering cost.
(iii) Find the optimal order to cover period and continue from
not optimum table for next order no.)
(iv) Substitute in the table
(v) Find the relevant cost )
Step 1:
40.0
EPP = = 80 (𝑐𝑎𝑛𝑡 𝑒𝑥𝑐𝑒𝑒𝑑 𝑓𝑟𝑜𝑚 𝑡ℎ𝑖𝑠 𝑜𝑝𝑡𝑖𝑚𝑎𝑙)
0.50
Step 2:

Order Order to cover Period Ordering Holding Cost

Number cost
1 40 0
(initial inventory)
1,2 40 1*(0.5)*(10) = 5
1
1,2,3 40 5 + 2*(0.5)*(20) = 25*
(optimal and less than EPP)
1,2,3,4 40 25 + 3*(0.5)*(50) = 100
4 40 0
4,5 40 20
2 4,5,6 40 20
4,5,6,7 40 35*
4,5,6,7,8 40 195
8 40 0
8,9 40 7.5
3 8,9,10 40 12.5
8,9,10,11 40 57.5*
8,9,10,11,12 132.5
4 12 0
 © Majid [email protected]
©inventory control_Lumpy model

Step 3:
Week Total
1 2 3 4 5 6 7 8 9 10 11 12
Demand 40 10 20 50 40 0 10 80 15 5 30 25
Order 70 100 130 25 4
Balance 30 20 0 50 10 10 0 50 35 30 0 0 235
Step 4:

Total relevant costs

(for 4 orders and 235 total inventory (from the above
table))
= $40 x 4 + 0.50*(235) = $277.50

(e) Least Unit Cost.

(*hint: (i) Find EPP = (Setup Cost)/(Holding cost)

(ii) Find unit order and total cost per unit (minimizes the per-
unit ordering and holding costs of each order)
(iii) Find the optimal order to cover period and continue from
not optimum table for next order no.)
(iv) Substitute in the table
(v) Find the relevant cost )
Step 1:
40.0
EPP = = 80 (𝑐𝑎𝑛𝑡 𝑒𝑥𝑐𝑒𝑒𝑑 𝑓𝑟𝑜𝑚 𝑡ℎ𝑖𝑠 𝑜𝑝𝑡𝑖𝑚𝑎𝑙)
0.50
 © Majid [email protected]
©inventory control_Lumpy model

Step 2:

Order Order to Ordering + Holdingcost Unit Total cost

Number cover Order per unit
Period (round
off)
1 40 40 1
(40/40=1)
1 1,2 40 +1*(0.5)*(10) = 45 50 0.9*
(45/50 =
0.9)
1,2,3 45 + 2*(0.5)*(20) = 65 70 0.93

3 40 20 2
3,4 65 70 0.93*
2 3,4,5 105 110 0.95
5 40 40 1
5,6 40 40 1
3 5,6,7 50 50 1*
5,6,7,8 170 130 1.31
8 40 80 0.5
4 8,9 47.5 95 0.5*
8,9,10 52.5 100 0.53
10 40 5 8
5 10,11 55 35 1.57
10,11,12 80 60 1.33

Step 3:
Week Total
1 2 3 4 5 6 7 8 9 10 11 12
Demand 40 10 20 50 40 0 10 80 15 5 30 25
Order 50 70 50 95 60 5
Balance 10 0 50 0 10 10 0 15 0 55 25 0 175
 © Majid [email protected]
©inventory control_Lumpy model

Step 4:

Total relevant costs

(for 5 orders and 175 total inventory (from the above
table))
= $40 x 5 + 0.50*(175) = $287.50