Modelling just in time purchasing in the ready mixed concrete industry 2

138

CHAPTER 5 TESTING THE ULTIMATE EOQ-JIT COST
INDIFFERENCE POINT MODEL

5.1 Introduction
Chapter 3 and Chapter 4 theoretically suggested that it was possible for an EOQ system
to be more cost effective than a JIT purchasing system when the inventory annual
demand was greater than its break-even point, even when the JIT operations could
experience or take advantage of inventory physical plant space reduction. A case study
relating to the procurement of cement in the ready mixed concrete (RMC) industry in
Singapore is presented in this chapter to empirically examine this proposition. The case
study was conducted in the cement division of RMC supplier
S
I in November 2003.

Section 5.2 describes the background of RMC supplier
S
I . Section 5.3 states the
assumptions (boundary conditions) for this test. Section 5.4 and 5.5 derive the ultimate
EOQ-JIT cost indifference points for the procurement of cement. Section 5.6 discuss on
the ultimate EOQ-JIT cost indifference points and the cost indifference points derived by
previous researchers. Section 5.7 performs sensitivity analyses to inform readers
concerning the application and limitations of models developed.

5.2 The background of RMC supplier
S
I
The group of companies in which RMC supplier
S
I was a subsidiary, was incorporated

in April 1973 and listed in the Singapore Stock Exchange in 1979. The business scope of
the group included the sale of RMC, dry mix mortar products and cement. The group also
139

had four cement manufacturing plants in China when the case study was carried out in
November 2003.

The cement division of RMC supplier
S
I in Singapore built two huge silos on the Pulau
Damar Laut Island and adopted an EOQ system to procure its cement from Japan. The
designed carrying capacity of each of the silos was approximately 25,000 tonnes. The
average area occupied by one silo was about 2,800
2
m
. The sum of the carrying
capacities of the two silos was approximately 40,000 tonnes, where safety and flexibility
factors had been considered. Stock flexibility parameter was
=
b
1.25. Cement imported
by the cement division of supplier
S
I
was mainly shipped using 40,000-tonne cement
carriers. Supplier
S
I
placed an order approximately once a month for about 40,000
tonnes of cement. The annual demand (

D
) was 520,000 tonnes in 2003.

The annual cost of carrying one tonne of cement was
=
h
S$ 321. The carrying cost can
be split into cement check-in cost (
checkin
h
), cement storage costs (
1storage
h and
2storage
h ) and
cement check-out cost (
checkout
h ). The cement check-in cost was the depreciation and
operating costs of the facilities to unload cement from a cement carrier to a silo and
$Sh
checkin
=
199 /year per tonne. The cement storage costs were
1storage
h
= S$ 8 / year per
tonne and
2storage
h = S$ 14 / year per tonne. The cement storage costs
1storage

h included
insurance, cement spoilage cost and opportunity cost of the working capital tied up with
the cement purchased. The cement storage costs
2storage
h included the depreciation cost of
the silo facilities, utilities, personnel salaries and property tax.
140

Mankiw (1997, p.6) defined the opportunity cost of an item as “what you give up to get
that item.” Potts (2002) suggested that based on the principle of opportunity cost, the
economic value of a resource was determined by its next best alternative use. Potts’
(2002) suggestion indicates that to compute the opportunity cost of the working capital
tied up with the cement purchased accurately, the alternative investment plans for the
amount of capital have to be worked out. This is however a difficult task. Nevertheless,
Heyne (1996) showed that the opportunity cost of the working capital tied up with an
inventory could be practically computed through such low-risk investments as
government bonds. Hence, the average interest rate of government bonds in Singapore in
the year 2003 was used to compute the opportunity cost of the working capital tied up
with the cement purchased.

The cement check-out cost was the depreciation cost and operating cost of the facilities,
mainly cement trucks, to deliver cement from the silo to a RMC batching plant and
checkout
h
=S$ 100 / year per tonne. Typical cement unloading facilities, storage facilities,
and delivery facilities are shown in Figures 5.1, 5.2 and 5.3 respectively. The cost of
placing an order was
=
k
S$ 432,000 / order for transportation alone. A 40,000-tonne

cement carrier is shown in Figure 5.4. Each tonne of cement took up
=
α
0.112
2
m
of the
inventory facility space. The annual cost to rent a square meter of inventory facility was
=
F
S$ 84. If cement was purchased under a JIT system in Singapore, the cost was =
J
P
S$ 69 / tonne.


141


(a) Cement unloading
facilities
(b) Cement transportation belt

Figure 5.1 Facilities at a cement bulk terminal



(a) Packing facility (b) Control panel

Figure 5.2 Cement storage facilities

Figure 5.3 Cement check out facility: a cement truck
142



Figure 5.4 A 40,000-tonne cement carrier

The order was often raised one or two months before the departure of a cement carrier
from Japan. The Japanese cement manufacturers offered a few alternative pricing
strategies. Two of the pricing strategies are discussed below. Pricing strategy 1 is suitable
for the EOQ without price discount system. Pricing strategy 2 is suitable for the EOQ
with a price discount system. Purchasing cement according to the pricing strategy 1 cost
E
P =S$ 42 / tonne. Under pricing strategy 2, the delivery price started at =
O
E
P S$ 45 /
tonne. For every additional tonne ordered, the price would decrease by $S
E
=
π
5
105.7
−
x

for the entire order lot. The discount could be valid for order quantity up to

=
max
Q
50,000 tonnes, when the price per unit became
=
min
E
P
S$ 41.25 / tonne. Beyond
this level, the price remained the same.

It is essential to note that the information for this case study was collected through
interviews with the overseas investment manager, the financial manager, the production
manager and the customer service supervisor of the cement division of supplier
S
I
in
November 2003. At this point, it should be noted that the examples given by Fazel (1997,
143

p.502), Fazel
et al
. (1998, p.107) and Schniederjans and Cao (2000, p.291; 2001, p.115)
were hypothetical.

It is also important to note that although the cement division of RMC
supplier
S
I
ordered its cement in an EOQ fashion, the accounting system adopted by

them did not exactly follow the EOQ approach. However, it was suggested that the cost
information could be structured as above to fit the EOQ model. The value of each
parameter was the average value.

5.3 Boundary conditions
To compare the present study with the study of previous researchers and to make the
problem simple, demand variability and safety stock were not considered in this case
study. In addition, it was found that each tonne of cement took up approximately 0.115
2
m
of the 100-tonne silo and approximately 0.109
2
m
of the 25,000-tonne silo. The
annual cost of holding one tonne of cement in a 100-tonne silo was slightly higher than
S$ 321 and the annual cost of holding one tonne of cement in a 25,000-tonne silo was
slightly lower than S$ 321 /year per tonne. This shows that the annual cost of holding one
tonne of cement in silos can roughly be assumed to be a constant. Each order of cement
was delivered by the 40,000 tonne cement carrier. Hence, the ordering cost under the
EOQ system can be assumed to be fixed per order.

Under the EOQ without price discount system, the optimal economic order quantity was
close to the routine order quantity. The annual inventory ordering cost item (
Q
kD
) (i.e.,
432,000 x 520,000 / 40,000), was S$ 5,616,000 / year. The annual inventory carrying cost
144

(

2
Qh
) was S$ 6,420,000 / year. This shows that the annual inventory ordering cost item
(
Q
kD
) was close to the annual inventory carrying cost item (
2
Qh
). Based on Eq. (3.2), the
economic order quantity was =
∗
Q 37,411 tonnes / order. Hence, the economic order
quantity (
∗
Q
) was close to the routine order quantity 40,000 tonnes / order.

Under the EOQ with a price discount system, the optimal economic order quantity was
also close to the routine order quantity. The routine cement order quantity for cement,
40,000 tonnes / order, was less than
max
Q , the maximum order quantity that can be
ordered and still receives a price discount at rate
E
π
under the EOQ with a price discount
model. Hence, the EOQ with a price discount system in the case study was actually a
max
QB

elowEOQd
system. Based on Eq. (4.3), the optimal economic order quantity was
=
∗
d
Q 42,998 tonnes / order. Hence, the optimal economic order quantity was close to the
routine order quantity.

Based on the above background, the assumptions of boundary condition 2, namely,
assumptions No.1 to No.2 and No. 4 to No. 9 in Table 1.1 can, thus, be roughly satisfied.
Therefore, the EOQ-JIT cost indifference points for cement purchasing in the cement
division of RMC supplier
S
I can be computed by the models developed in Chapters 3
and 4. It should be noted that the additional costs and benefits resulting from JIT
purchasing are not considered in this case study.

145

5.4 Ultimate EOQ-JIT cost indifference point under the EOQ without price discount
system
Eq. (3.24) and Eq. (3.25) can be used to derive the break-even points under the EOQ
without price discount system. Eq. (3.26) can be used to derive the ultimate EOQ-JIT
cost indifference point under the EOQ without price discount system. According to Eq.
(3.24), the inventory facility break-even point was 4,644
2
m
. According to Eq. (3.25),
the annual demand break-even point was 408,830 tonnes. Based on Eq. (3.26), the
ultimate EOQ-JIT cost indifference point represented by the annual demand in the

cement division of RMC supplier
S
I was 414,557 tonnes. Since a) the floor area of the
two silos, 5600
2
m
, was greater than the break-even point, 4,644
2
m
, and b) the ultimate
EOQ-JIT cost indifference point, 414,557 tonnes, was greater than the break-even point,
408,830 tonnes; therefore, the value of the ultimate EOQ-JIT cost indifference point
under the EOQ without price discount system was confirmed to be 414,557 tonnes.
According to Eq. (3.17), the annual carrying capacity of the two silos was 594,444 tonnes,
which was capable of accommodating 520,000 tonnes of cement. The annual carrying
capacity of these two silos under the EOQ without price discount system can be as high
as 928,818 tonnes, which is substantially greater than the annual demand in 2003, if the
flexibility parameter
b
is set to be 1.

If the annual cost of holding one unit of cement,
h
, in Eq. (3.7) was replaced by the
cement storage cost,
1storage
h
, this equation was then converted to be the formula for
computing the EOQ-JIT cost indifference point proposed by Fazel (1997, p.499).
146

According to Fazel’s (1997, p.499) model, the EOQ-JIT cost indifference point for
cement purchasing should have been 9,481 tonnes / year.

Each tonne of cement occupied at least 0.1
2
m
of the silo. Hence, JIT purchasing of
cement could have taken advantage of inventory physical plant space reduction. Based on
the models proposed by Schniederjans and Cao (2001, p.116), when “saving space and
using it to house additional increasing amounts of inventory to meet larger annual
demand are juxtaposed issues … a JIT system would virtually always be preferable to an
EOQ system”. Hence, the EOQ-JIT cost indifference point would be
∞
+
. The EOQ-JIT
cost indifference points, worked out with the models proposed by Fazel (1997),
Schniederjans and Cao (2001) and the author, are shown in Table 5.1.
Table 5.1 A comparison of the EOQ-JIT cost indifference points under the EOQ without
price discount system
Fazel’s (1997)
model
Schniederjans and
Cao’s (2001) model
Author’
model
Cement purchasing
EOQ-JIT cost
indifference point
9,481

(tonnes / year)
∞
+

(tonnes / year)
414,557
(tonnes / year)


5.5 Ultimate EOQ-JIT cost indifference point under the EOQ with a price discount
system
Eq. (4.31) and Eq. (4.19) can be used to derive the break-even points under
the
max
QB
elowEOQd
system. Eq. (4.36) can be used to derive the ultimate EOQ-JIT cost
indifference point under the
max
QB
elowEOQd
system. Setting
∗
d
Y in Eq. (4.31) to zero and
solving it with Matlab, the inventory facility break-even point under
the
max
QB
elowEOQd

system (
∗
eqd
N
) was worked out to be 5,270
2
m
. Substituting
E
N
with
147

this amount in Eq. (4.19), the annual demand break-even point under
the
max
QB
elowEOQd
system (
∗
eqd
D
) was worked out to be 422,518 tonnes. The Matlab code
and the figure of the difference between the function of the annual carrying capacity of
the inventory facility and the function of the EOQ-JIT cost indifference point,
∗
d
Y
, with
respect to

E
N
are attached in Appendix 4. Based on Eq. (4.36), the ultimate EOQ-JIT
cost indifference point under the
max
QB
elowEOQd
system was 424,322 tonnes. The upper
limit of the silo size that can still reap the benefit of
E
π
was
max
bQ
α
= 7,000
2
m
. The
upper limit of the annual demand that could still reap the benefit of
E
π
was
kQ
h
Q
E
22
2
max

2
max
+
π
= 647,700 tonnes. Hence, the value of the ultimate EOQ-JIT cost
indifference point under the
max
QB
elowEOQd
system was confirmed to be 424,322 tonnes.
This was because a) the floor area of the two silos, 5600
2
m
, was less than the upper
limit 7,000
2
m
and greater than the break-even point, 5,270
2
m
; and b) the computed
ultimate EOQ-JIT cost indifference point, 424,322 tonnes, was above the break-even
point, 481,271 tonnes, and less than the upper limit, 647,700 tonnes; According to Eq.
(4.27), the annual carrying capacity of the two silos was 465,217 tonnes, which was
capable of accommodating 424,322 tonnes of cement. The annual carrying capacity of
these two silos could be as high as 647,700 tonnes, which was substantially greater than
the annual demand in 2003, if the flexibility parameter
b
was set to 1.

If the annual cost of holding one unit of cement,
h
, in Eq. (4.18) was replaced by the
cement storage costs,
1storage
h , this equation was then converted to be the formula for
148

computing the EOQ-JIT cost indifference point proposed by Fazel (1998, p.106).
According to the model of Fazel et al. (1998, p.106), the EOQ-JIT cost indifference point
for cement purchasing was 9,796 tonnes / year.

Schniederjans and Cao (2000, p.294) argued that a JIT ordering system was preferable to
an EOQ system at any level of annual demand and with almost any cost structure, the
EOQ-JIT cost indifference point proposed by them thus should be
∞
+
. The EOQ-JIT
cost indifference points, worked out with the models proposed by Fazel, et al. (1998),
Schniederjans and Cao (2000) and the author, are shown in Table 5.2.
Table 5.2 A comparison of the EOQ-JIT cost indifference points under the EOQ with a
price discount system
Model of Fazel et
al. (1998)
Author’s
model
Schniederjans and Cao’s
(2000) model
Cement purchasing
EOQ-JIT cost

indifference point
9,796
(tonnes / year)
424,322
(tonnes / year)
∞
+

(tonnes / year)


5.6 Discussion
The batching capacity of the widely used batching plant was 90
3
m
/ hr in Singapore. The
average demand for cement of the 90
3
m
/ hr batching plant was approximately 40,500
tonnes / year in 2003 as estimated by the production manager of the RMC batching plant
division of RMC supplier
S
I . Based on the surveys presented in Chapter 2, the numbers
of batching plants owned by RMC suppliers in Singapore were arranged from one to
seventeen. Hence, the annual cement demand of each RMC supplier surveyed was at
least 40,500 tonnes / year. This figure was significantly greater than the EOQ-JIT cost
indifference point worked out from the models proposed by Fazel (1997) or by Fazel et al.
(1998). Hence, the EOQ-JIT cost indifference point derived from the models of these
149

researchers suggested that all the RMC suppliers should operate in an EOQ fashion.
However, RMC supplier
S
A ,
S
B ,
S
C ,
S
H ,
S
J ,
S
K ,
S
L ,
S
M ,
S
N and
S
O were
purchasing their cement in a JIT fashion. These RMC suppliers used a number of 100-
tonne silos to store their buffer stock, as shown in Figure 5.5. The 100-tonne silos were
filled on a daily basis. Hence, the EOQ-JIT cost indifference point derived from Fazel’s
(1997) model and the model of Fazel et al. (1998) were not supported by the information
on cement purchasing in the RMC industry in Singapore. At the same time, the EOQ-JIT
cost indifference point models proposed by Schniederjans and Cao (2000, 2001)
suggested that all the RMC suppliers should operate in a JIT fashion, as cement

purchasing can take advantage of physical plant space reduction. However, the cement
division of RMC supplier
S
D ,
S
E ,
S
F ,
S
G , together with
S
I were purchasing their
cement in an EOQ fashion. This is shown in Table 2.6. To reap economies of scale, the
cement division of these RMC suppliers built a number of huge multi-cells silos on the
Pulau Damar Laut island, as shown in Figure 5.6. The cement received at the Pulau
Damar Laut Island was then delivered to their RMC batching plant divisions and
batching plants of other RMC suppliers. Hence, cement purchasing in the RMC industry
in Singapore did not support the EOQ-JIT cost indifference point proposed by
Schniederjans and Cao (2000); rather it supported the one developed in this study.

It is important to highlight the economies of scale in cement storage. As stated earlier, a
representative tonne of cement takes up approximately 0.115
2
m
of floor area in a 100-
tonne silo and approximately 0.109
2
m
in a 25,000-tonne silo. In addition, in terms of the
overall throughput, the annual cost of holding one tonne of cement in a 100-tonne silo is

150

approximately S$ 330 /year per tonne, which is slightly above S$ 321 /year per tonne;
while the annual cost of holding one tonne of cement in a 25,000-tonne silo is
approximately S$ 312 /year per tonne, which is slightly below S$ 321 /year per tonne.
The difference in the construction costs of many small silos as opposed to one large silo
should also be addressed. However, the construction cost of a silo has already been
considered as a component of the depreciation cost of the silo facilities. The annual cost
of holding one tonne of cement in a cement silo is calculated based on the property tax,
insurance, cement spoilage cost, opportunity cost of the working capital tied up in the
purchased cement, the depreciation cost of the silo facilities, utilities, personnel salaries,
and the depreciation cost and operating cost of the facilities to unload cement from a
cement carrier to a silo. The annual cost of holding one tonne of cement in a 100-tonne
silo is close to that of a 25,000-tonne silo, as bulk cement must be stored in silos that are
waterproof, clean and protected from contamination, dry (internal condensation
minimized) and with stocks rotated in chronological order of the dispatch dates marked
on delivery documents (Zacharia, 1985; Singapore Productivity and Standards Board,
1986; Mao, 1997). The fact that the annual cost of holding one tonne of cement in a
25,000-tonne silo lies slightly below S$ 321 /year per tonne can shift the actual EOQ-JIT
cost indifference point to be lower than 414,557 tonne / year (for the EOQ without a price
discount system) or 424,322 tonne / year (for the EOQ with a price discount system).

Furthermore, the actual EOQ-JIT cost indifference point could be modified to an even
lower value, if the out-of-stock cost was considered. On the other hand, the EOQ-JIT cost
indifference may shift to be a greater value if the impact of inventory policy on quality
151

and flexibility were considered. This will be further discussed in Chapter 6 and Chapter 7.
It is also important to note that the case study also suggested that the annual carrying
capacity of an inventory facility dropped from 594,444 tonnes to 465,217 tonnes when a

price discount rate
5
105.7
−
x
was offered, where the flexibility parameter was 1.25. The
reason for this reduction in annual carrying capacity has already been explained in
Section 4.4.5.1.

The EOQ models assume that the demand of an inventory is known and fixed. Hence, the
optimal economic order quantity is fixed. However, the annual demand of an inventory in
practice is seldom a constant. The company may have difficulties to rent additional
inventory facility when the annual demand of the inventory increases. In such a case, the
inventory order frequency can be increased to match the increased annual inventory
demand, and the inventory order size may remain the same as the routine order size. This
suggests that the carrying capacity of an inventory facility in practice can be greater than
“the annual carrying capacity of an inventory facility”, thus again making it possible for
an inventory facility to hold the EOQ-JIT cost indifference point’s amount of inventory.

One important reference quoted by Schniederjans and Cao (2001) to support their
argument that JIT was virtually always the preferable alternative for inventory purchasing
decisions was the study conducted by Pan and Liao (1989). In Pan and Liao’s (1989)
study, the EOQ model was converted into a series of JIT purchasing models that could be
used to determine inventory deliveries and cost savings, and demonstrated that there was

152



Figure 5.5 100-tonne cement silos

Figure 5.6 25,000-tonne cement silos

153

no limitation on the cost advantage of using JIT, based on the model parameter of annual
demand. This raises the question whether it was economical to use 2,500-tonne cement
carriers, rather than 40,000-tonne cement carriers, to conduct frequent deliveries. This
question was raised to the production manager of the cement division of supplier
S
I . The
production manager explained that the transportation of bulk Portland cement must use
specialized transportation vehicles, such as cement trucks or cement carriers, as shown in
Figure 5.3 and Figure 5.4. The transportation cost of bulk Portland cement from Japan to
Singapore by a 40,000-tonne cement carrier was about S$ 10.8 / tonne. The transportation
cost of bulk Portland cement from Japan to Singapore by a 2,500-tonne cement carrier
was about S$ 20.0 / tonne. The transportation cost of bulk Portland cement in Singapore
by a cement truck was as high as S$ 0.3 / tonne per kilometer. In addition, as indicated by
E
π
the purchase price could be increased if cement was ordered in small lot sizes. The
difference between the selling price,
J
P and the purchase price,
O
E
P , was only S$ 24. The
average delivery cost of cement was around S$ 4.0 / tonne in Singapore, where it was

assumed that the average transportation distance was between 10 and 20 kilometers,
because Singapore is a relatively small island. In addition, the expensive operating and
depreciation costs of the cement silos and cement check-in facilities must be paid. To
sum up, it was not economically justifiable for the cement division of supplier
S
I to split
its order size from 40,000 to 2,500 tonne to match the available cement carriers and to
deliver in a JIT pattern.



154

5.7 Sensitivity analyses
Sensitivity analyses were carried out to determine how the ultimate EOQ-JIT cost
indifference point models were affected by variations in the parameters in the models.
The analyses were used to identify parameters on which more attention should be
concentrated when selecting cement purchasing approaches. Following Kometa et al.
(1996), Ling (1998) and Schniederjans and Cao (2001), sensitivity analyses were
restricted to the major parameters only, so as to limit the complexity of the results. These
parameters were 1) the price difference between the JIT purchasing system and the EOQ
system (
EJ
PP − ) or (
0
EJ
PP − ), 2) the annual cost of carrying one unit of inventory in
stock (
h
), 3) the cost of placing an order (

k
), 4) the annual cost to own and maintain a
square meter of physical plant space (
F
), and 5) the price discount (
E
π
).

The steps for undertaking the sensitivity analysis, following Schniederjans and Cao
(2001), are given below:
•

Step 1: The ultimate EOQ-JIT cost indifference point was computed in a normal way
using data given in Section 5.2 for the ultimate EOQ-JIT cost indifference point
model (see Eq. (3.26) and Eq. (4.36)). This step was performed in Sections 5.4 and
5.5.
•

Step 2: The value of the first parameter was varied from -10% to +10% and the
percentage change in the ultimate EOQ-JIT cost indifference points was computed.
•

Step 3: Step 2 was repeated for the remaining parameters to compute the percentage
change in the ultimate EOQ-JIT cost indifference point.

155

The sensitivity analyses for cement purchasing by the cement division of RMC supplier
S

I were conducted under the EOQ without price discount system and EOQ with a price
discount system.

Table 5.3 shows the change in the ultimate EOQ-JIT cost indifference points when the
changes were made to the parameters individually under the EOQ without price discount
system. The ultimate cost indifference points were computed by using Eq. (3.26).
Table 5.3 Percentage change in the ultimate EOQ-JIT cost indifference point under the
EOQ without price discount system

Percentage change in the cost indifference point Percentage
change in
parameter
(
EJ
PP −
), holding
J
P
as a constant
h

k

F

-10% 22.5% -9.2% -9.2% -0.8%
-5% 10.4% -4.6% -4.6% -0.4%
0 0 0 0 0
5% -8.9% 4.6% 4.6% 0.4%
10% -16.7% 9.2% 9.2% 0.8%

Table 5.4 shows the change in the ultimate cost indifference points when the changes
were made to the parameters individually under the EOQ with a price discount system.
The ultimate cost indifference points were computed by using Eq. (4.36).

Table 5.4 Percentage change in the ultimate EOQ-JIT cost indifference point under the
EOQ with a price discount system

Percentage change in the cost indifference point Percentage
change in
parameter
(
0
EJ
PP −
), holding
J
P
as a constant
h

k

F

E
π

-10% 17.5% -9.3% -7.6% -0.7% 1.9%
-5% 8.3% -4.6% -3.8% -0.4% 0.9%

0 0 0 0 0 0
5% -7.4% 4.6% 3.7% 0.4% -0.9%
10% -14% 9.3% 7.3% 0.7% -1.8%
156


The implications of Tables 5.3 and 5.4 are three-fold. First, the changes in the cost
indifference points were not linearly related to the changes with the parameters. This is
explicit in Eqs. (3.26) and (4.36). Second, the price factor was the most sensitive
parameter for cement purchasing in the cement section of RMC supplier
S
I Tables 5.3
and 5.4 show that the ultimate EOQ-JIT cost indifference point changes the most when
the value of the purchase prices were varied. Third, among the parameters listed in Tables
5.3 and 5.4, the rental was the least sensitive parameter. This is not unexpected, because
rental was only a component of the physical storage costs for cement storage. The major
components of the physical storage costs, for example, cement check in-facilities, cement
check-out facilities, personnel salaries, etc. were considered in
h
in the models
developed in this study
.


5.8 Summary
Chapter 5 is dedicated to a case study in the RMC industry in Singapore which showed
that it is possible for the EOQ system to be more cost effective than the JIT system, when
the annual demand is greater than the ultimate EOQ-JIT cost indifference point, even
when the JIT operation can take advantage of inventory physical plant space reduction.
The case study also suggests that this conclusion can be valid only if the order quantity

under the EOQ system cannot be economically split.

As suggested in Chapter 1, the intention of Chapters 3 and 4 was to theoretically examine
the capability of an inventory facility to hold the EOQ-JIT cost indifference point’s
amount of inventories based on the mathematical models developed by previous
157

researchers. Hence, the additional costs and benefits of EOQ or JIT purchasing were not
considered in the models developed. The additional costs and benefits of the EOQ and
JIT purchasing of cement in the RMC industry in Singapore may be balanced by each
other. Hence, the EOQ-JIT cost indifference point models developed in the previous
chapters were still well supported by the data on cement purchasing in the RMC industry
in Singapore even though the additional cost components were not considered in models
developed in these chapters. However, these additional cost components may not always
be balanced by each other.

Based on the studies conducted by other researchers, for example, the studies of Rao and
Sheraga (1988), Johnson and Stice (1993), Cheng and Podolsky (1996), Low and Chan
(1997), Low and Choong (2001d), Singh (2003), Low and Wu (2005a, b), Wu and Low
(2005a, b, c, d) and others, the impact of inventory purchasing policy on quality and
production flexibility and out-of-stock costs should be considered in the EOQ-JIT cost
difference models. In addition, the models developed in Chapter 3 and Chapter 4 were
general models, rather than particularly designed for the RMC industry. Hence, Chapter 6
will consider these additional cost components and examine how these additional cost
components may affect the selection of inventory purchasing policy in the RMC industry.