BACKORDER RAW MATERIAL INVENTORY CONTROL SYSTEM

WITH LEAD TIME AND ORDERING COST REDUCTION: CASE

STUDY PT ICI PAINTS INDONESIA

Tiena Gustina Amran; Dinar Suryo Lesmono

Industrial Engineering Department, Faculty of Industrial Engineering, University of Trisakti

Jln. Kyai Tapa No. 1, Grogol, Jakarta Barat 11440
tiena_amran@yahoo.com; tiena_inovasi@trisakti.ac.id1; dinar.lesmono_ie02@yahoo.com2

ABSTRACT

This research investigates the continuous review inventory model involving variable lead
time with partial backorders, where the amount is uncertain. PT ICI Paints Indonesia is one of
chemical industries which produced paint. The main problems faced by company are often occur
stock out and also over stock of raw material replenishment, and the second case is raw material
ordering lead time which can’t be controlled and determined its extent so the company more
depend on the supplier policy. The purpose of this research is to suggest the optimal raw material
inventory system alternatives synchronized with the stock out characteristics and the condition can
be occurred. Besides, the backorder raw material inventory control can reduce lead time and raw
material ordering cost. After the calculation, the optimal solution of inventory models can be
obtained; those are with backorder condition which produced the annual inventory total cost
amount of 15.069.845.628, 27 rupiahs; which mean it just effectively save inventory cost by 5, 81%
for the company.

Keywords: inventory, cost reduction, lead time, back order

ABSTRAK

Penelitian ini menginvestigasi keberlanjutan model tinjauan inventori yang

mengikutsertakan variable masa tenggang dengan backorder parsial yang jumlahnya tidak jelas.
PT ICI Paints Indonesia adalah satu dari banyak industri kimia yang memproduksi cat.
Permasalahan utama yang dihadapi perusahaan adalah seringnya kehabisan atau kelebihan stok
pergantian bahan baku serta masa tenggang pemesanan bahan baku tidak bisa diatur dan
ditentukan sehingga perusahaan bergantung pada aturan penyuplai. Tujuan penelitian ini untuk
menyarankan sistem inventori bahan baku alternatif yang sesuai dengan karakter dan kondisi stok.
Selain itu, pengaturan backorder inventori bahan baku bisa mengurangi masa tenggang dan biaya
pemesanan bahan baku. Setelah dikalkulasi, solusi optimal dari model inventori berjumlah total
15.069.845.628,27 rupiah, yang berarti secara efektif menghemat biaya inventori perusahaan
sebesar 5,81%.

Kata kunci: inventori, pengurangan biaya, lead time, back order

94 INASEA, Vol. 12 No.2, Oktober 2011: 94-106

 INTRODUCTION

Extend (Q, R) Inventory Model Under Lead Time and Ordering Cost Reduction by Wu &
Lin (2004) is a probabilistic model of Continuous Inventory Review (Q, R) including variable of
lead time which can be used in Backorder and lost sales conditions. The main purpose of using this
model is to having the optimal solutions simultaneously, including the amount of the economic
order, decrease the order costs, reordering point and also decreased lead times, and minimize total
inventory costs.

Inventory model is very important because there are many benefits come from the
application of this model (Tersine, 1994). When the company’s applying the JIT (Just In Time)
concept is very precise and give priority for the short lead times in delivery of raw materials or the
production process schedule (Ramasesh, 1993). Shorten the lead time is become the most effective
and efficient way to get the application of JIT goals.

The main inventory purpose of the JIT program is producing small lot sizes in good quality
(Tersine, 1994). In order to achieve this goal, in reducing the order costs or the preparation costs in
capital invested is one of the most effective ways. Moreover, produce the smaller lot sizes because
the order cost or preparation costs lower, there are many advantages on it, such as; scheduling
flexibility, smaller warehouse, and lower investment inventory (Sohal & Naylor, 1992).

Based on the PT ICI Paints Indonesia, Cikarang Plant problems, the final task of this research
aims to provide some alternative suggestions inventory control optimal system for the classification
of raw material. This adapted to the characteristics and existing conditions in stock out which is,
the Backorder and lost sales condition and can reduce lead time and cost of raw material delivery
also.

Theoretical Background
The inventory of raw materials is divided in 2 alternatives, first by using the model of
Continuous Review (Q, R), and the second with the Extend model (Q, R). Steps using this model
found in Figure 1. The steps are: (1) getting past demand data and then used as input models. Past
demand data that has nature probabilistic (uncertain); (2) calculation process is performed by
processing data requests and data inventory costs; (3) from the calculation process can be generated
Q, R, SS, B, and the optimal total cost of supplies each year. Extend model (Q, R) can be optimal
produced by S and L; (4) after calculating, have the production resumed; (5) as a feedback from
production to meet demand, and then the expected demand data will be fulfilled and the possibility
of stock out can be minimized.

Notation and Assumptions

A very important assumption in this model is the lead time which is comprised of several
components, namely ordering inventory, transit reservations, and lead times from suppliers,
delivery time and set up time. In practice, the lead time can be reduced by adding additional costs
in other words lead time can be controlled. By reducing the lead time we can reduce safety stock,
eliminating loss of profit due to stock out, improve customer satisfaction levels and also increase
the competitive advantage in a business environment. The model of this inventory is certainly
involving shortages or stock out condition is good in a Backorder or lost sales.

Explanation and assumptions developed assumptions in this model can be described as

follows:

Backorder Raw Material… (Tiena Gustina Amran; Dinar Suryo Lesmono) 95

Q = number of orders (decision variable)
S = ordering costs / order (decision variable)
R = re-order point (decision variable)
L = length of lead time (decision variable)
D = average demand /year
h = cost savings / unit / year
π = stock out costs, backorder, or lost sales / kg
π0 = profit / kg
β = possibility of stock out, backorder, or lost sales, 0 ≤ β ≤1
π = π + (1 − β)π 0
S0 = first order cost
I(S) = large financial investment required to achieve cost S, 0 < S ≤ S 0
θ = opportunity cost of investment / unit time
δ = percentage of messages in which the cost of S / Rp. decreased at the time of investment
I(S)Increased
xL = demand during lead time
σL = standard deviation of demand during lead time
X = demand during lead time with the average x L And standard deviation σ L , when σ is
standard deviation demand per unit of time
Y = number of items received, is a random
variable
ψ(k) = {φ (k) - k[1 - Φ(k)]} = g(k)
(from the standard normal table) φ ,
Φ = standar normal probability density
function (p.d.f) and d.f
(from the standard normal table)
α = bias factor ( 0 ≤ α ≤ 1 ) when the number of items received less than or equal with the
number of orders Q.

Assumptions
There are assumptions the researchers had before the research. Firstly, that the reordering
point R = demand during lead time + safety stock (SS), dan SS = k x (standard deviation of demand
during lead time), can be explained, R = x L + k σ L , where k is the safety factor.

Secondly, inventory is continuously reviewed and re-ordering occurs when inventory levels
fall just at the point position reorders R. Thirdly, lead time L has n free components. Components -
i have a minimum duration ai and normal duration bi, crashing cost per unit time ci. To simplify, ci
is assembled as c1 ≤ c 2 ≤ ... ≤ c n . It is very clear that the decrease in lead time was first performed
on component 1 (for having a crashing cost the most minimum), and continued component 2, and
so on.

Fourthly, if we set L0 = ∑n bj and Li i as a long lead time with components 1, 2, ..., i, i is

j=1

changed to its minimum duration, then Li can be expressed as Li = ∑ j=1 bj − ∑ j=1 (bj − aj) , i = 1, 2, ...,
n i

96 INASEA, Vol. 12 No.2, Oktober 2011: 94-106

n ; and therefore the lead time crashing cost R (L) per cycle for L ∈ [Li, Li-1], which can be
i −1
explained by the following equation: R(L) = ci(Li − 1 − L) + ∑ cj(bj − aj)
j=1

Fifthly, it is assumed that the investment of money, I (S), which serves to reduce the cost of
messages is logarithmic function of the cost of the message S. This can be explained in the
following equation: I(S) = b ln ⎛⎜ S0 ⎞⎟ with 0 < S ≤ S 0 , where b = 1 .
⎝S⎠ δ

Sixthly, variance of the number of items received are: Var(Y | Q) = σ 02 + σ12 Q 2 . In this
study, it is assumed σ 02 = σ 12 = 0 and α = 1 , This shows that the number of items received equal to
the number of orders (Q).

Basic Model
Within this model, it can be determined: the cost of S is a variable decision and the
decision sought to minimize the cost of investment capital to reduce the order cost of S and
inventory costs by optimizing the Q, S, R, and L with limits 0 < S ≤ S 0 . Therefore, the purpose of
the problems mentioned above is to minimize the total cost of this inventory as mentioned below:

EAC(Q,S,R,L) = θI(S) + EAC(Q, R, L) =

⎛ S ⎞ SD
θb ln⎜ 0 ⎟ + [
+ h R − x L + (1 − β)E(X − R) + ] +
h
[
σ 02 + (σ12 + α 2 )Q 2 ] +
πD
E(X − R) +
⎝ S ⎠ αQ 2αQ αQ
R(L)D (2-27)
+ , for 0 < S ≤ S0 .
αQ

It is assumed that demand during lead time X follows a normal distribution with mean -
average x L and standard deviation σ L . The equation is R = x L + k σ L and
∞ ∞
E(X-R)+ = ∫ (x − R)dF(x) = ∫
R k
σ L (z − k)dΦ(z)
= σ L {φ (k) - k[1 - Φ(k)]}
= σ Lψ(k)
= σ Lg(k) ,
when φ and Φ is the standard normal probability density function (pdf) and df, therefore we can
determine the safety factor k as a decision variable in addition to reordering point R. Thus, the
equation that can explain all these issues, for backorder conditions are:
min EAC(Q,S,R,L) =
⎛ S ⎞ SD + [kσ L + (1 − β)σ L ψ(k)]
θb ln⎜ 0 ⎟ + +
h
[σ 02 + (σ12 + α 2 )Q 2 ] +
πD
σ L ψ(k)
⎝ S ⎠ αQ 2αQ αQ
+
R(L)D
, for 0 < S ≤ S . (2-28)
αQ
0

To determine Q, we first determine

∂EAC(Q, S, k, L) 2
= − SD − hσ 0 + h (σ 2 + α 2 ) − πD σ L ψ(k) - R(L)D (2-30)
1
∂Q 2
αQ 2
2α Q 2α 2
αQ αQ 2
∂EAC(Q, S, k, L) =
∂S

Backorder Raw Material… (Tiena Gustina Amran; Dinar Suryo Lesmono) 97

 − θb D (2-31)
+
S αQ
∂EAC(Q, S, k, L) = hσ L − h(1 − β)σ L (1 − Φ (k))
∂k
πD (2-32)
− σ L (1 − Φ(k))
αQ
∂EAC(Q, S, k, L) = 1 −1 1 −1
hkσ L 2 + h(1 − β)σL 2 ψ(k)
∂L 2 2
1 πD − 1 2 D (2-33)
+ σL ψ(k) − ci
2 αQ αQ

For the final solution of the Q, S, and k, the expected total cost of inventory each year will
be happen in final position of the interval [Li, Li-1]. On the other hand, for a given value
of L ∈ [L i , L i −1 ] by determining the equation (2-30), (2-31), and (2-32) is zero, it can be
determined:

⎡ h ⎤
2D⎢S + σ02 + πσ Lψ(k)+ R(L)⎥
Q= ⎣ 2D ⎦ (2-34)
h(σ12 + α 2 )
αθbQ (2-35)
S=
D
and
hα Q (2-36)
Φ (k) = 1 −
h(1 − β) αQ + D π
(For backorder conditions)

METHOD

Research framework of this final task research can be seen in the Figure 1 below.

Figure 1 Framework Chart Research

98 INASEA, Vol. 12 No.2, Oktober 2011: 94-106

Description Figure 1:
Q = Number Booking Economic
R = Point Reservations Back
SS = Safety Stock
USL = Service Level
S = Order Costs
L = Lead Time

From Figure 1, the framework shows that inventory system used by companies today
compared with the proposed system of raw material inventory control.

RESULT AND DISCUSSION

Calculation Process of Continuous Review Inventory Model (Q, R) Backorder

Conditions
Continuous Review Model (Q, R) Backorder is an inventory model used when the
company has a stock out of consumer’s demand, so they do the fulfilment delay of the demand and
immediately made an emergency order, the other hand, and consumers still waiting until the goods
are available.

Because the usage of raw material data has normal distribution, so the calculation will use
the Standard Normal Distribution Table in the Appendix.

RE 5875 Raw Material

P1 = Rp. 8.182,00 / kg
S = Rp. 200.000,00 / order
h1 = Rp. 736,38 / kg / year
π 1 = Rp. 818,20 / kg
L = 5 days
x1 = 26.075,149 kg
D1 = x 1 (12) = (26.075,149)(12)
= 312.901,7820 kg
σ1 = 8.473,327 kg
r = Lead time = 5 days = ,2273
Demand Period 22 days
Standard deviation of demand during lead time
= σ L1
= σ1 r
= (8.473,327)( 0.2273 )
= 4.039,4999 kg
The average usage during the lead time
= x L1
= x1 r
= (26.075,149)(0.2273) = 5.926,1701 kg

Iteration 1 (no stock out)

Q1 * = 13.037,1608 kg

Backorder Raw Material… (Tiena Gustina Amran; Dinar Suryo Lesmono) 99

OSOR1 = 0,0375
k = z0,0375 = 1,78
(from the Standard Normal Distribution Table)
g(k) = g(1,78) = 0,015
(From the Stand. Normal Dist. Table)
USOR1 = 0,0046 (the optimal stock out probability)
Iteration 2
Q1 * = 14.563,646 kg
g(k) = 0,0166
k = z0,0166 = 1,74
Iteration 3
Q1 * = 14.717,1291 kg
g(k) = 0,0168
k = z0,0168 = 1,73
Iteration 4
Q1 * = 14.736,2021 kg
g(k) = 0,0168
k = z0,0168 = 1,73
Iteration 5
Q1 * = 14.736,2021 kg
g(k) = 0,0168
k = z0,0168 = 1,73

Due to the value of Q1 * and k at iteration 4 and 5 are equal then the iteration is stopped.

Further calculations:
SS1 = 6.988,3348 kg
R1 = 12.914,5049 kg
B1 = 67,8636 kg
m1 = 21,2335 times
T1 = 1 / (21,234)
= 0,0471 years = 12 days
USL1 = 1 – 0,0046 = 0,9954

Conclusion of the calculations of Continuous Review (Q, R) Backorder inventory model

for raw materials is the 5875 RE:

1. Q1 = 14.736,2021 kg
2. R1 = 12.914,5049 kg
3. B1 = 67,8636 kg
4. SS1 = 6.988,3348 kg
5. m1 = 21,2335 times ≈ 22
6. T1 = 0,0471 years = 12 hari
7. USL1 = 0,9954 = 99,54 %
8. TC1 = Rp. 2.576.159.894,83

Continuous Review (Q, R) Inventory Model Backorder Calculation Result

Summary
The Continuous Review (Q, R) Inventory Model Backorder calculation result for all the
classification of raw materials shown in Table A1.

100 INASEA, Vol. 12 No.2, Oktober 2011: 94-106

Percentage Decrease in the USL for Raw Materials that R Value exceed Q

The Table 1 shows that there are material items that have an R value exceeding Q, namely
WH 0118. To overcome this distortion the R value would be reduced by lowering the percentage of
the USL to obtain R values of less than Q.

The formula used in this calculation for the – j raw materials is:

1. Unit Stock out Rate :

USORj = 1 - USLj
2. Partial Expectation :
Qj USORj
g(k) =
σLj
3. Reorder Point :
Rj = x Lj + SSj = x Lj + k σLj

While other variables such as the calculation of R, B, SS, M, T, USL, and TC using the same
formula as that used in Continuous Review (Q, R) Backorder Inventory Models.

WH 0118 Raw Material (j=2)

Based on the input data and the calculated Q 2 in inventory models Continuous Review (Q,
R) Backorder, then the input data of this calculation is:

P2 = Rp. 18.315,00 / kg
S = Rp. 2.000.000,00 / order
h2 = Rp. 1.648,35 / kg / year
π2 = Rp. 1.831,5 / kg
x L2 = 15.776,9875 kg
σ L2 = 6.582,7490 kg
D2 = 138.837,4896 kg / year
Q2 = 21.860,0394 kg
USL2 = 97,91 % ( first USL )
USL2 = 97 % (Enter by trial and error)
USOR2 = 0,03
g(k) = 0,0996
k = 0,9 (from the Standard Normal Distribution Table)
R2 = 21.701,4616 kg < Q2 (21.860,0394 kg)
So the percentage of the new USL 2 is 97%.
B2 = 655,6418 kg
SS2 = 5.924,4741 kg

The results of model calculations Continuous inventory Review (Q, R) Backorder for the
classification of raw materials after a decline in the USL A shown in Table 2. In Table 2, Q has
experienced rounding tailored to the individual - their raw materials. For example, raw materials
RE 5875, Q = 15,000 kg, derived from: 15 (number of places) multiplied by 1000 kg (capacity
place (tanks)). As a result of this rounding TC adjust to change Q.

Calculation Process for Extend Inventory (Q, R) Backorder Conditions

Iteration algorithm

Backorder Raw Material… (Tiena Gustina Amran; Dinar Suryo Lesmono) 101
 In the step 1, for every Li, i = 0,1,2,...,n, do (i)-(v). Beginning with Si1=S0 and ki1=0
(resulting ψ(k) i1 = 0,3989, which can be determined by checking the normal table φ (k i1 ) = 0,3989
and Φ(k i1 ) = 0,5). Substitution S i1 and ψ(k) i1 to equation (3), calculate the Qi1.

Noted that by using Q i1 S i2 set of equations (4). By using Q i1 specify Φ (k i2 ) from

equation (5). Then found by examining the table, and also obtained ψ(k) i2 . Repeat (ii) through (iv)
until no change in the value of Q i, S i, and k i.

The step 2 is comparing S i and S 0. If S i <S 0, then the solution has been found in step 1
were optimal for given L i. Determine the solution by (Qi*, Si*, ki*). If Si ≥ S0, then for Li, use the
Si* = S0, and (Qi*, ki*) can be determined by substitution Si* to equation (2-33), and solve (3) and (5)
by iteration until there is convergent (the same solution obtained by Step 1).

The step 3 for each (Qi*, Si*, ki*, Li), i = 0, 1, 2, ..., n, calculate the total cost of inventory
every year is expected EAC(Qi*, Si*, ki*, Li) using equation (1) for Backorder conditions.

The step 4 is to discover min EAC(Qi*, Si*, ki*, Li). If EAC(QN, SN, kN, LN) = min
i =0,1,2,...,n i =0,1,2,...,n
EAC(Qi*, Si*, ki*, Li), then (QN, SN, kN, LN) is the most optimal solution. Please note, if k N and L N
has been determined, then the reordering point RN = x LN + kN σ LN will follow. Then we find the
Continuous Review (Q, R) Inventory Model Backorder Conditions Result Calculations for 21 raw
materials as results.

In the step 5, we find Continuous Review (Q, R) Inventory Model Backorder Result
Calculation after the Reduction of USL (Results Rounding). Total Lot = QJ before rounding (kg) /
Minimum Lot Size (kg), the results in-Roundup.

Raw Material RE 5875

The raw material data and cost data of RE 5875 necessary for the calculation are as
follows:

x = 26.075,149 kg
D = 312.901,7820 kg
σ = 8.473,327 kg
r = 0,2273
xL = 5.926,1701 kg
σL = σ r = 4.039,4999 kg
P = Rp. 8.182,00 / kg
S = Rp. 200.000,00 / order
h = Rp. 736,38 / kg / year
π = Rp. 818,20 / kg
π0 = 35 % purchase price
= Rp. 2.863,70
L = 5 days

σ 02 = 0
2
σ 1 = 0
α = 1

102 INASEA, Vol. 12 No.2, Oktober 2011: 94-106

 θ = 1
δ = 1/b
b = 2.500.000
β = 0 dan 1
ci = 30 % purchase price
= Rp. 2.454,60
ci = 50 % purchase price
= Rp. 4.091,00

Table 1 Lead Time Data for RE 5875 Raw Materials for Extend (Q, R) Backorder Inventory Model

Unit
Lead Time Normal Minimum
Crashing
Component Duration Duration
Cost
(i) bi (days) ai (days) ci (Rp/days)
Manufacturing
2 1 2454.6
LT (1)
Transportation
3 2 4091
LT (2)

Iteration 1

Step 1 : Calculate the lead time Li for each β, where β = 0 and β =1.
L0 = 5 days
L1 = 4 days
L2 = 3 days

Step 2 : Calculate the lead time crashing cost R (L) for each β.
R(L0) = 0
R(L1) = Rp. 2.454,60
R(L2) = Rp. 6.545,60

Step 3 : Calculate the cost of stock out per cycle.

For β = 0 :
π = Rp. 3.681,90
For β = 1 :
π = Rp. 818,20

Step 4 : Set Si1 = S0 and k Si1 = 0 (leading to g (ki1) = 0.3989 and OSOR = 0.5, which can be
determined by looking at the standard normal table).

Step 5 : substitution value of Si1 and g(ki1) into the Q equation to calculate the Qi1.

For β = 0 :
L0 = 5 days
Q01 = 72.193,7034 kg
L1 = 4 days
Q11 = 68.423,1145 kg
L2 = 3 days
Q21 = 63.882,8475 kg

For β =1:

Backorder Raw Material… (Tiena Gustina Amran; Dinar Suryo Lesmono) 103
L0 = 5 days
Q01 = 35.922,1915 kg
L1 = 4 days
Q11 = 34.266,6346 kg
L2 = 3 days
Q21 = 32.301,9577 kg

Iteration 2

Step 62 : Based on calculations on the Qi1, Si2 set by using the equation for S.

For β = 0 :
L0 = 5 days
S02 = Rp. 576.808,0237
L1 = 4 days
S12 = Rp. 546.682,0456
L2 = 3 days
S22 = Rp. 510.406,5491

For β = 1 :
L0 = 5 days
S02 = Rp. 287.008,5245
L1 = 4 days
S12 = Rp. 273.781,0758
L2 = 3 days
S22 = Rp. 258.083,8426

Step 72 : Using the results of calculations that have been Qi1 obtained above, determine OSORi2
from equation OSOR, and then determine ki2 and g(ki2) obtained from standard normal tables.

For β = 0
L0 = 5 days
OSOR02 = 0,9559
k02 = 1,7
g(k02) = 0,0183)
L1 = 4 days
OSOR12 = 0,9581
k12 = 1,73
g(k12) = 0,017
L2 = 3 days
OSOR22 = 0,9608
k22 = 1,73
g(k22) = 0,0158

For β = 1
L0 = 5 days
OSOR02 = 0,8967
ki2 = 1,26
g(k02) = 0,0495
L1 = 4 days
OSOR12 = 0,9014
k12 = 1,29)
g(k12) = 0,0465)

104 INASEA, Vol. 12 No.2, Oktober 2011: 94-106

L2 = 3 days
OSOR22 = 0,9071
k22 = 1,32
g(k22) = 0,0436

Step 82 : Ssubstitute the value from Si2 and g (ki2) into the Q equation to calculate Q i2.

For β = 0 :
L0 = 5 days
Q02 = 26.860,7367 kg
L1 = 4 days
Q12 = 25.668,3957 kg
L2 = 3 days
Q22 = 24.372,4775 kg

For β = 1 :
L0 = 5 days
Q02 = 19.569,0325 kg
L1 = 4 days
Q12 = 18.750,3750 kg
L2 =3
Q22 = 17.881,6266 kg

Iteration 3

The calculating process on the iteration 3 and the next iteration using the step - the same
steps listed at iteration 2 to get the optimal results when there was no change in the value of Qi, Si,
and ki. It is continue to step 10.

Step 10: The final stage in the process of this calculation is to calculate the total cost of
inventory every year is expected to EAC (Q i, S i, k i, Li) for each lead time and β.

For β = 0 :
L0 = 5 days
EAC (Qi,Si,ki,Li) = Rp. 2.577.014.269,6586
L1 = 4 days
EAC (Qi,Si,ki,Li) = Rp. 2.568.646.208,5280
L2 = 3 days
EAC (Qi,Si,ki,Li) = Rp. 2.575.306.210,1497

For β = 1 :
L0 = 5 days
EAC (Qi,Si,ki,Li) = Rp. 2.575.437.790,0866
L1 = 4 days
EAC (Qi,Si,ki,Li) = Rp. 2.574.794.520,0706
L2 = 3 days
EAC (Qi,Si,ki,Li) = Rp. 2.574.107,734,9293

After that followed by calculating safety stock SS, the number of stock out B, frequency of
order m, the interval ordering T, USL and the level of service.

SS = 5.976,3579 kg
B = 33,793 kg

Backorder Raw Material… (Tiena Gustina Amran; Dinar Suryo Lesmono) 105
m = 32,0408 times
T = (1 / (32,0408))(264) = 8 days
USL = 0,9965

Table 2 Summary of Extend (Q, R) Inventory Model Backorder Conditions for RE 5875 Raw Material

Li R(Li) Si* Q i* Ri* EAC*(Qi*,Si*,ki*,Li)

β
(days) (Rp) (Rp) (kg) (kg) (Rp)
5 0 74.756,99 9.356,6387 16.105,7099 2.577.014.269,66
0 4 2.454,60 74.454,93 9.318,8315 13.845,7933 2.568.646.208,53
3 6.545,60 74.857,10 9.369,1686 11.409,4498 2.575.306.210,15
5 0 79.336,39 9.929,7991 13.601,2199 2.575.437.790,09
1 4 2.454,60 78.124,63 9.778,1339 11.641,8397 2.574.794.520,07
3 6.545,60 78.025,40 9.765,7144 9.532,0599 2.574.107.734,93

CONCLUSION

The output generated from this inventory model is number of economic order (Q) and the
inventory position must be booked at the optimal raw material (R), the number of possible stock
out (B), the amount of safety stock (SS), and the optimal service level (USL).

The total cost of inventory each year (TC) was obtained Rp.15.069.845.628, 27 for the
model Continuous Review (Q, R) Backorder, and Rp.15.061.749.314, 98 to Extend (Q, R)
Inventory Model Under Lead Time Cost reductions and Ordering Backorder conditions. The
company produced today is for Rp.16.000.000.000, 00. This means that savings have occurred for
5.81% compared with the model of Continuous Review (Q, R) and for Backorder 5.86% when
compared to (Q, R) Inventory Model Under Lead Time and Ordering Cost reductions Backorder
conditions.

Also, it reduced cost of messages (S) and lead time (l) sending any materials that can
reduce the total cost of supplies each year to Extend (Q, R) Inventory Model Under Lead Time and
Ordering Cost reductions Backorder conditions.

REFERENCES

Tersine, R. J. (1994). Principles of inventory and materials management. New Jersey: Prentice
Hall.

Ramasesh, R. V. (1993). A logistics-based inventory model for JIT procurement, International

Journal of Operations & Production Management, 13 (6), 44-59.

Sohal, A. S., & Naylor, D. (1992). Implementation of JIT in a Small Manufacturing Firm,
Production and Inventory Management Journal 33(1), 20-27.

Wu, K. S., & Lin, I. C. (2004). Extend (r,Q) inventory model under lead time and ordering cost
reductions when the receiving quantity is different from ordered quantity. Quality &
Quantity 38: 771 – 786.

106 INASEA, Vol. 12 No.2, Oktober 2011: 94-106