SCIENTIFIC

GLASS, INC.:
INVENTORY MANAGEMENT
Candas Demir
Onur Yilmaz
Burcu Yüzüak
January 2010, Ankara
I. INTRODUCTION
In this case study, production and operations management (POM) issues of a mid-s
ize company,
named as Scientific Glass Inc., in a highly growing market are studied. Using th
e background
information on past actions of the company to correct inventory management and t
heir results, and
considering the market leadership opportunity, how inventory management approach
can be made
better is explained by evaluating different alternatives from different aspects.
In the first part, critical
POM issues are mentioned, following that these problems are analyzed. In the thi
rd part, alternative
options are listed and then they are evaluated. Finally, considering the trade-o
ffs of these evaluations, a
conclusion is made. And it must be mentioned that, throughout the case, related
points are referenced to
the case text and lecture notes with corresponding page and paragraph numbers.
II. CRITICAL POM ISSUES
As mentioned in the text, there is an identified increasing trend in the balance
s of inventory
levels. (Page 1, para. 2) For a growing company in a growing market, this high i
nventory level, in other
words tied up money in the inventory, creates an obstacle for this company to us
e this extra capital on
other areas, such as expansion to international markets. Also, as mentioned, deb
t to capital ratio
exceeded the target level of 40% and with the same approach this increase of thi
s ratio also jeopardizes
the company s funding expansion plans to international markets. Although, there ar
e many other POM
issues are found in the text, these mentioned two were the most critical ones an
d it is thought that if
they are solved the other problems will be solved spontaneously.
III. ANALYSIS OF THE CRITICAL POM ISSUES
In the last part, it is mentioned that average inventory level is high enough to
jeopardize
company s future plans. Therefore, main reasons behind this problem should be anal
yzed. First of all,
company has a policy related to 99% fill rate, which is also open to discussion
considering the market
average of 92%, and warehouse managers are usually exceed even this limit and th
ey are keeping more
inventory than necessary (Page 4, para. 5). Secondly, company has a policy to no
t to exceed 60 day s
supply, which is also open to discussion, and most warehouse managers are exceed
ing this upper limit
(Page 5, para. 6). Considering all these aspects, it is found that inventory lev
els and transshipment costs
should be decreased and at the same time responsiveness to customer should be in
creased in order to be
a market leader. By doing these, simultaneously, approach of the warehouse manag
ement could be
changed to a better position by changing policies related to them as it is tried
in the past with different
ways and failed (Page 6, para. 6). In addition, when this inventory level kept u
nder control, debt to
capital ratio will be saddled since extra capital tied up in the inventory will
be available to be used.
IV. ALTERNATIVE OPTIONS FOR PROBLEMS
In order to solve the analyzed problems in the previous part, there are actually
two main aspects
to consider: firstly, number of warehouses and their structure can be changed; s
econdly, related policies
can be changed and of course appropriate ones can be done simultaneously.
For changing the number of warehouses, in other words, centralizing or decentral
izing warehousing
functions, available options are considered as:
.
Continuing with 8 warehouses: This option makes no change on the network of the
warehouses and all regions will be supplied its warehouse if there is no stock-o
ut occurs.
.
One central warehouse: In this option, one central warehouse near to manufacturi
ng facility at
Waltham will send all customer orders from one location.
.
Two centralized warehouses: In this option, addition to the main warehouse at Wa
ltham, there
will be an additional warehouse at the west, at Phoenix, and it will be supplied
from Waltham.
Demand of east region will be met from Waltham, demand of west region will be me
t from
Phoenix and demand of central region will be met from both warehouses, assuming
to have
equal shares on the central region.
.
Outsourcing the warehousing functions: In this option, all warehousing actions w
ill be
outsourced to Global Logistics (GL) and distribution will start from main wareho
use at
Waltham and then GL will be responsible from rest of the operations.
In addition to these options, there are some policy change proposals which try t
o make POM
approach better, like periodic audits and increasing reporting activity levels,
stopping trunk stock
activities etc. Since these policy changes can be applied at different warehousi
ng functions these
proposals will be analyzed one by one and their possible effects will be conside
red.
V. EVALUATION OF ALTERNATIVE OPTIONS
Evaluation of mentioned alternatives will be conducted from mainly five aspects:
transportation
costs, average inventory levels, time responsiveness, fill rates and finally add
itional costs and benefits.
1-) Transportation Costs: Transportation costs for alternatives are calculated f
or the two products,
namely Griffin and Erlenmeyer, since they are mentioned as the best representati
ve for a total of nearly
3000 products of Scientific Glass (Page 2, para. 3). In addition, for each optio
n, demand for the next
year calculated considering the 20% increase in sales (Page 8, para. 3). When wa
rehouse to customer
shipments are considered average shipment weight of 19,5 pound is used and to ha
ve an average
transportation cost value, these two products costs are averaged according to the
ir relative proportion
in sales (Page 8, para. 3). It also be mentioned that, inter-warehouse transship
ments occur only when
stock-out occurs and as the number of warehouses are decreasing, effect of this
costs will be
diminished; therefore, it is only considered in the option where there are 8 war
ehouses.
For the first option, having 8 warehouses and making no change, from Waltham to
all other 7
warehouses all items are sent by bulk shipment. Inter-warehouse transshipments a
re calculated by bulk
shipment rates and they are considered only when a stock-out occurs, therefore f
ill rate is included in
these calculations. Finally, for every warehouse, customer shipment costs are ca
lculated as all tabulated
in Appendix Table 1 and average total cost found as $2701,41.
For the second option, when there is only one warehouse, all customer shipments
are calculated
for rates of Winged Fleet and results are given in Appendix -Table 2. In this op
tion, average total cost
is calculated as $12210,16.
For the third option, when two centralized warehouses considered, it is assumed
that Waltham
will supply east region, Phoenix will supply will west region and they will equa
lly supply the central
region. With this approach, transportation costs are calculated in Appendix Tabl
e 3 and average total
cost is found as $2332,07.
For the last option, when warehousing functions are outsourced, assuming the 5 r
egions of
Global Logistics (GL) will have equal amount of demand, as tabulated in Appendix
Table 4, total
average cost is calculated as $2276,83.
To conclude, as it is expected, when number of warehouses are decreased transpor
tation costs
are increased as can be seen from the Table 1 below. From the aspect of transpor
tation costs, GL option
has the smallest cost amount.
8 Warehouses 1 Warehouse 2 Warehouses Outsourcing
2701,41 12210,16 2332,07 2276,83
Table 1: Average Total Annual Transportation Costs
Gathered from Appendix Table 1-2-3-4
2-) Average Inventory Levels: First of all, it must be decided which inventory p
olicy that the
company should apply. Begin with the review type; although firm monitors all the
inventory transfers
from Waltham warehouse to other warehouses; they think taking physical counts of
inventory at all
warehouses (Page 6, para. 6). Therefore, it is concluded that company uses perio
dic inventory review
policy. Secondly, company did not mention any due date, therefore the inventory
plans should consider
infinite time horizon. And lastly, although there exists a fixed cost for shipme
nts from warehouses to
customers; there is no other fixed cost related to transportation to the warehou
ses, i.e. no fixed ordering
cost. The only order cost is $0.40 per pound bulk shipment cost which is a varia
ble cost with weight.
As a result, all analysis can be conducted considering critical ratios and the r
elated fill rate values,
which is the only option that is left and also it is considered as the most appl
icable to the situation.
Since some of the simultaneous changes can be done, considering ceteris paribus
principle and
when fill rate is maintained exactly as 99% for all warehouses, we can calculate
the average inventory
level that must be kept at warehouses. As shown in the Appendix -Table 5, for ha
ving 1, 2 and 8
warehouses average inventory levels are calculated for two representative produc
ts. When outsourcing
option is used, it will be the same for the company in the sense of kept invento
ry levels for the one-
centralized-warehouse option therefore they are assumed to be equal. Weighted-av
erage biweekly
levels are found as:
8 Warehouses 2 Warehouses 1 Warehouse Outsourcing
988,53 680,34 597,03 597,03
Table 2: Weighted-average biweekly inventory levels
Gathered from Appendix -Table 5

As shown in the Table 2 above, as demand aggregates, in other words, number of w

arehouse
decreases, level of inventory decreases as it is expected. This is because, the g
reater the degree of
collaboration, the lower the uncertainty (standard deviation of the error or coe
fficient of variation) of
the demand model (Lecture 9, Slide 6). This implies that the money tied up in the
inventory decreases
and this extra capital can be used in other areas, like expansion plans to inter
national markets.
3-) Time Responsiveness: Delivery system of the company compensates 2 weeks of s
hipment
cycles including the stock-out situations. In order to be a market leader, diffe
rentiation on this subject is
also needed and unfortunately since this is not an exact quantitative scale, onl
y possible situations
could be mentioned. For having one centralized, or two centralized or 8 decentra
lized warehouse
options, they all include at most 3 days ready to shipment duration (Page 6, par
a.2) and Winged Fleet s
delivery time of at most 3 days (Page 12, Exhibit 5) if there is no stock-out si
tuation and the stock-out
probabilites are diminishing with the aggregated demands. On the other hand, GL
has 1-day premium
shipment in addition to 3-day regular shipments (Page 8, para. 1). Considering t
he highly growing
market situation and different segment of products, having different delivery ti
mes to different products
and also to different customers will make this company focus on the most yieldin
g areas. Therefore, it
can be said that working with GL has the advantage of differentiating customers/
orders and, since
there will be 2 warehouses, stock-out probability and related durations will be
less compared to other
options. And all of these aspects will increase the time responsiveness of the c
ompany.
4-) Additional Costs and Benefits: Other qualitative and quantitative issues rel
ated to options
will be covered in this section. Firstly, in the text, it is mentioned that in o
rder to continue with the
current warehouses total of $10M investment is necessary (Page 5, para. 5), it i
s assumed that all of this
amount will be equally shared among all warehouses. Secondly, since warehouse op
erating costs will
be the %15 of the total warehoused inventory (Page 5, para.3), these costs could
be directly compared
with the annual average inventory levels that are kept in each option which are
calculated in the
previous section. Thirdly, the amount paid to sales forces (Page 3, para. 5) wil
l not change when the
company have 1, 2 or 8 warehouses because it is assumed that as the number of wa
rehouses decreased,
number of salesperson per warehouse will increase and total of 32 will not chang
e. On the other hand,
when warehousing is outsourced this amount will not be paid.

1 Warehouse 2 Warehouses 8 Warehouses Outsourcing

Warehouse
Inventory Level: Inventory Level: Inventory Level: Inventory Level:
Operating Costs:
597,3 x 26 =
15529,8
680,34 x 26 =
17688,84
988,35 x 26 =
25697,1
597,3 x 26 =
15529,8
Table 3: Additional Quantitative Comparisons
Replacing worn
equipment cost:
$10M/8 = $1,25 M $10M/8 x 2 = $2,5 M $10 M $10M/8 = $1,25 M
Sales Force
Payments:
32 x $33.000 +
Commision
32 x $33.000 +
Commision
32 x $33.000 +
Commision
No costs
As can be seen from the Table 3 above, decreasing number of warehouses will decr
ease the
replacing worn equipment costs and since demand aggregates, warehousing operatin
g costs will
decrease also. It is obvious that outsourcing will relieve the company from the
amount paid to the sales
forces.
In addition to these, there are some qualitative issues that must be mentioned.
First of all, when
GL is used for warehousing, as mentioned in the text (Page 8, para. 2), SG s senio
r managers will be
able to focus on increasing sales, marketing issues and developing next generati
on of products.
Secondly, there are some issues that must be mentioned from the proposed policy
changes. Stopping
the practice of trunk stock could conclude with a decrease in the time responsiv
eness and therefore it
should not be stopped. Thirdly, also as mentioned in the same proposed policy ch
anges, improving the
controlling systems will create a better understanding of the current situation
after the warehousing
functions changed. Finally, when GL is used, the approach of warehouse managers
to keep more than
99% fill rate and 60-day-supply will not be a problem, because all of these oper
ating issues will be
responsibility of GL. This will help to company not to keep excessive amount of
inventory and less
tied-up money in the inventory which can be used in other areas.
5-) Fill Rate: Company s fill rate policy should also be questioned. It is said th
at company
replaced the earlier fill rate policy of 93%, which is only marginally better th
at the industry average fill
rate of 92%, with 99% (Page 4, para. 1). However, there is no sign that the comp
any is implementing
this policy because it is the best approach that must be taken for the company o
bjectives. Moreover,
using a fill rate higher than optimal level leads to higher inventories and more
money tied up in the
inventory. Therefore, company should lower the rates down to optimal levels, if
there is no other
concern related to market leadership or customer satisfaction.
To calculate the optimal levels of fill rates for all four alternatives we shoul
d consider cost
items which are added to underage and overage costs. However, the sales leadersh
ip noted that
underage costs are 10% of the gross margin and overage costs are 0.6% of the uni
t cost of any product
(Page 4, para. 2). Also it is assumed that unit costs covers all the costs such
as warehouse rental and
operation costs, cost of capital and inventory write-offs. For the three alterna
tives other than
outsourcing, there is no change in cost items, only the multiplied quantities ar
e changed; but the
outsourcing alternative eliminates the 15% warehouse rental and operating costs
and 1% inventory
write-offs as calculated in the Appendix -Table 6. As a result, overage costs ar
e decreased while
underage costs are increased. Resulting optimal fill rates are as follows:
1, 2 or 8 Warehouses Outsourcing Outsourcing

Griffin

95.4%

96.5%

Erlenmeyer 94.9% 96.1%

Table 4: Optimal fill-rates for alternatives -Gathered from Appendix Table 6

These numbers can be interpreted in two different ways: First, if company is fle
xible about the
determination of fill rate, in other words if it can lower the fill-rates from 9
9% to optimal levels,
outsourcing option pushes the optimal fill rates to higher levels which results
in larger inventories and
more money to tie up. Second, if the company still insists on keeping fill rate
at 99%, the additional
costs that must be paid to maintain 99% fill-rate level is lowered in the outsou
rcing alternative.
Consequently, the better policy related to fill rates depend on the attitude of
the company.
Finally, another policy change about fill-rates can be considered. Rather than u
sing one fill-rate
for over all products of the company, different rates for different products can
help the company in
decreasing inventory costs related to, at least, for some of the products.
VI. CONCLUSION
To conclude, since available options are studied from different aspects, it must
be mentioned
that the company should choose the alternatives and compare the results of evalu
ations according to
their priorities. For instance, evaluation criteria like inventory levels and tr
ansportation costs are
conflicting on interests. Company can see their situation from an exchange curve
like in the below
graph (Graph 1) and make decisions according to priorities. The curve shows the
inventory and
transportation cost levels as the number of warehouses changes.

Graph 1: Exchange curve for inventory level and transportation cost

While exchange curves like given above can be used for the pairwise comparisons,
weighted
score model can be useful for an overall assesment of options (Lecture 3, Slide
26). A sample weighted
score model with hypotetical weights and scores is provided in the Table 5 below
, in order to derive a
what-if scenario study:
O P T I O N S
Weight 1 Warehouse 2 Warehouses 8 Warehouses Outsourcing
Transportation Costs 1 3 2 4
30 Average Inventory Levels 4 2 1 4
Time Responsiveness 3 2 1 4
15 Fill Rate 3 3 3 2
Additional Costs
5 Replacing Worn Equipment 4 3 2 4
Warehouse Operating 4 3 2 4
5 Sales Force Payments 3 3 3 4
Additional Benefits
5 Extra time of senior managers 2 2 2 4
Changes in warehouse management 1 1 1 4
Total: 100
TOTAL SCORES: 285 245 175 370

1: Poor 2: Acceptable 3: Good 4: Excellent

Table 5: Hypotetical Weighted Score Model
While assessing the weights for factors, it is considered that average inventory
level and the
transportation costs are the most important costs for the company. Then, the fil
l rate follows them and
it is assumed that the first interpretation of fill rates mentioned in the previ
ous part is used by the
company. Time responsiveness is the next important factor which is followed by a
dditional costs and
benefits with equal weights for each. Changes in warehouse management is conside
red as options other
than outsourcing does not provide radical policy changes which could make wareho
using management
better. These weights and the scores related to our previous investigations yiel
d that the outsourcing the
warehousing function to Global Logistics is the best alternative among all.
All of investigations and cost studies conducted in this case study are to find
the most cost
effective option in order to getting closer to the target debt to capital ratio
of the company and provide
more capital to fund expansion into new international markets while maintaining
or even improving the
high customer satisfaction level.
VII. APPENDIX
Table 1: Calculating Transportation Costs for 8 Warehouses Option
Griffin Erlenmeyer Comments
Total Demand 65,04 19,56 Calculated by 20% increase
Fill Rate 0,01 0,01 If there is no policy change about it.
Bulk Shipment Rate 0,4 0,4

Inter-Warehouse Transportations (If stockout occurs)

Total Biweekly Costs0,0325 0,0196
Fill Rate x Total Demand x Cost
Rate x Weight
Total Biweekly Cost for 8 w/h
0,2601 0,156
Total Annually Costs (1) 6,7641 4,068

From Waltham to Warehouses

Total Amount to Carry 455,28 136,92

To seven other warehouses
Total Amount to Carry * Weight *
Bulk Rate
Total Annually Costs (2) 591,864 355,992
Total Biweekly Cost 22,764 13,692
From Warehouses to Customers (Average 19,5 pounds shipments by Winged Fleet)

Total Amount to Carry 520,32 156,48

Total Biweekly Cost 92,123 55,41
Fixed Cost + Total Demand x Cost
Rate x Weight
Total Annually Costs (3) 2395,20 1440,66
Total Annual Costs (1+2+3) 2993,83 1800,72
Relative Weight in Sales 75,49% 24,51%
Related Sales Revenue / Total
Sales Revenue from both
Average Annual Cost 2701,4

Table 2: Calculating Transportation Costs for 1 Centralized Warehouse Option

Griffin Erlenmeyer Comments

Total Demand (Biweekly) 520,08 156,36
Demand for one w/h region
65,01 19,545 Calculated by dividing total demand by 8
Demand for Central Region 130,02 39,09 To Toronto and Chicago
Demand for West Region
195,03 58,63 To Seattle, Denver and Phoenix
Demand for East Region 130,02 39,09 To Atlanta and Massachusetts
Total Costs for Within Region 154,99 46,60 For East Region
Total Costs for Across 1 Region 160,825 48,35 For Central Region
Total Costs for Across 2 Region 246,234 74,03 For West Region
Total Biweekly Cost 562,05 168,979
Total Annual Cost
14613,38 4393,46
Relative Weight in Sales 75,50% 24,50%
Related Sales Revenue / Total Sales
Revenue from both
Average Annual Cost
12109,16
Table 3: Calculating Transportation Costs for 2 Centralized Warehouses Option Tr
ansportation Costs for 2 Centralized Warehouses Option
Griffin Erlenmeyer Comments
Total Demand (Biweekly) 520,08 156,36 Calculated by considering 20% increase
Demand for one w/h region 65,01 19,545 Calculated by dividing total demand by 8
Demand for Central Region 130,02 39,09 To Toronto and Chicago
Demand for West Region 195,03 58,635 To Seattle, Denver and Phoenix
Demand for East Region 130,02 39,09 To Atlanta and Massachusetts

[ (Demand of West * Weight) ]* (5 x 1/ 19,5 +

Phoenix's Delivery Costs (1) 48,957 29,438
1.16 ) + [Demand of Central * 0,5 * Weight]
*(12 x 1/ 19,5 + 1.16)
[ (Demand of East * Weight) ]* (5 x 1/ 19,5 +
22,516
Waltham's Delivery Costs (2) 37,447
1.16 ) + [Demand of Central * 0,5 * Weight]
*(12 x 1/ 19,5 + 1.16)
From Waltham to Phoenix

Total Amount to Carry 260,04 78,18 West's Demand + Half of Central's Demand
Total Cost (Biweekly) (3) 13,002 7,818 Weight x Amount x Bulk Shipment Rate

Total Costs (Biweekly)

99,407 59,7726
(1+2+3)
Total Annual Costs 2584,58 1554,09
Related Sales Revenue / Total Sales Revenue
Relative Weight in Sales 75,50% 24,50%
from both
Average Annual Cost
2332,076

Table 4 : Cost Calculation for Outsourcing the Warehouse

Bulk Shipment Costs to Atlanta
Griffin Erlenmeyer Comments
Total Demand 13521,6 4066,8 Calculated by 1.2 x Demand of 2009
Bulk Shipment Cost Rate 0,4 0,4
Total Demand * Weight * Bulk Shipment Cost
Rate
Delivery Costs from Atlanta
Total Annual Costs (1) 676,08 406,68
Griffin Erlenmeyer
Total Demand
13521,6 4066,8
Demand for one region 2704,32 813,36 It is assumed five regions have equal deman
d.
Demand of Region * Weight * / Average
Southeast Region Costs 289,327 87,0191 Shipment Weight * Related Cost (from Exhi
bit
5)
Northeast Region Costs 327,812 197,1877
Central Region Costs 385,712 232,0162
Southwest Region Costs 424,370 255,267
Northwest Region Costs 443,612 266,845
Total Annual Costs (2) 1870,835 1038,337
Total Annual Costs (1+2) 2546,915 1445,0175
Related Sales Revenue / Total Sales Revenue
Relative Weight in Sales 0,75489 0,245
from both
Average Annual Cost 2276,834
Table 5: Calculating biweekly overall inventory levels for options
Griffin Erlenmeyer
8 W/h 2 W/h 1 W/h 8 W/h 2 W/h 1 W/h
Mean (2009) 54,2 216,7 433,4 16,3 65,2 130,4
Variance (2009) 21,4 38,3 51 10,9 19,5 26
Mean (2010) 65,04 260,04 520,08 19,56 78,24 156,48
Variance (2010) 30,816 55,152 73,44 15,696 28,08 37,44
Biweekly Ranges
Lower -14,157 118,29936 331,3392 -20,77872 6,0744 60,2592
Upper (Q*) 144,237 401,78064 708,8208 59,89872 150,4056 252,7008
Overall biweekly stock level 1153,897 803,56128 708,8208 479,18976 300,8112 252,
7008
Z(alpha) = Z(0.01) -2,57
Relative Weight in Sales Griffin: Erlenmeyer: Related Sales Revenue / Total Sale
s Revenue from
0,7549 0,2451 both

8 Warehouses

2 Warehouses

1 Warehouse

Weighted Average
988,526 680,337 597,026
Inventory Levels
Table 6: Calculation of fill rates
1-2-8 Warehouse Options Outsourcing
Griffin Erlenmeyer Griffin Erlenmeyer Comment
Unit Cost 3,96 4,56 3,3264 3,8304 Taken from Exhibit 3
Unit Price 8,8 9,5 8,8 9,5 Taken from Exhibit 3
Insurance Cost 0,0396 0,0456 0 0 %1 of the unit cost
Cost of Capital 0,554 0,638 0,465696 0,536 %14 of the unit cost
Warehousing Operations 0,594 0,684 0 0 %15 of the unit cost
Gross Margin 4,84 4,94 5,4736 5,6696
Underage Cost 0,484 0,494 0,54736 0,5667 10% of gross margin
Overage Cost 0,0238 0,0273 0,01995 0,02298 0.6% of unit cost
Fill Rate 0,9532 0,9475 0,9648 0,9610