Using System Dynamics to Estimate
Reductions in Life-Cycle Costs
Arising From Investments in Improved Reliability
William R. Killingsworth
Executive Director, MIT Forum for Supply Chain Innovation
Massachusetts Institute of Technology
77 Massachusetts Avenue, Room 1-176
Cambridge, Massachusetts 02139
billk@ mit.edu
Director, Office for Enterprise Innovation and Sustainability
University of Alabama in Huntsville
301 Sparkman Drive
Huntsville, Alabama 25899
256.824.4434
william.killingsworth@ uah.edu
Stephen M. Speciale
Nelson T. Martin
Office for Enterprise Innovation and Sustainability
University of Alabama in Huntsville
301 Sparkman Drive
Huntsville, Alabama 35899
256.824.2681
stephen.speciale@ uah.edu
martinn@ uah.edu
Abstract
“Doing more with less” has become a long-running and recurring theme across the
globe. Affordability is now a key metric for operations and sustainability, and reliability
is now seen as a key driver of these lifecycle costs. A system dynamics model has
been developed of an aviation supply chain that enables evaluation of alternative cases
in which investments are made to improve reliability, lower total demands, and reduce
spending on new procurement and overhaul over the lifecycle. It is shown that the
payback potential of an investment depends upon annual demand for the part, cost of
the part, percent improvement in reliability achieved, and any increase in cost of the part
due to the re-design. The analysis show that returns can be high and payback periods
can be fast, particularly for investments to improve reliability of items with high demand
and high cost. The research also indicates that close coordination is needed between
program management, procurement planning and acquisition in order to fully realize
savings. Ongoing research is developing reliability investment strategies and estimates
for lifecycle costs under differing demand, manufacturing and overhaul scenarios.
This research was conducted at the University of Alabama in Huntsville.
Introduction
“Doing more with less” has become a long-running and recurring theme across the
globe. Companies, government agencies and even charities are being forced to deliver
higher performance with reduced funding and capital. (Ain, 2009; Anthony, 2009; Shute,
2009; Gottlieb, 2008) This challenge is especially acute for the US Department of
Defense and the branches of the armed services where demands are great and
budgets are tight. As long ago as 1995, Dr. Paul Kaminsky, then the Under Secretary
of Defense for Acquisition and Technology, stated that a key goal “...is a simple one of
trying to do more with less.” This objective has steadily become more critical over the
last fifteen years and has led to on-going efforts to achieve efficiencies while at the
same time maintaining availability and system readiness. Because typically the costs to
operate, maintain, and dispose of a weapon system account for about 72 percent of the
total cost of ownership, much effort has focused on these expenditures. (GAO-03-57)
The DoD Reliability, Availability, Maintainability, and Cost Rational Report Manual
(2009) states the issues succinctly:
The Department of Defense (DoD) needs to acquire reliable and maintainable
products that are of high quality and readily available to satisfy user requirements
in meeting mission capability and operational tasks. The Department must
acquire these products at the most reasonable cost to the taxpayer. The cost to
the government, however, is not just computed by the procurement costs, but
also must balance the long-term costs incurred in maintenance, driven by
reliability, availability, and other factors throughout the system life cycle.
Improvements in reliability have multiple cost saving impacts such as fewer parts to buy
and overhaul, smaller inventories of replacement parts, fewer inspections, reduced
maintenance hours and down time, reduced transportation costs to ship replacement
parts, and on and on. Nevertheless, recent research found the following to be the case:
Test results since 2001 show that roughly fifty percent of DoD’s programs are
unsuitable at the time of initial operational test and evaluation, because they do
not achieve reliability goals. This represents a significant and alarming change in
the number of programs found unsuitable as compared to historical levels.
Because reliability is a prime determinant of long-term support costs, delivered
reliability so far off the mark has serious consequences for both operational
suitability and affordability. (Forbes, Hees, Long, and Stouffer, 2007)
Because of the strong tie between reliability and sustainment costs, the DoD Director of
Operational Test and Evaluation sponsored research to investigate the empirical
relationships between reliability investments, improvements in reliability and life-cycle
support costs. In this research, a preliminary relationship between investment in
reliability (normalized by average production unit cost) and achieved reliability
improvement was developed. This relationship is presented in Figure 1 and is taken
from a presentation delivered by Mr. Charles E. McQueary, Director, Operational Test
and Evaluation.
Figure 1: Empirical Relationship Between Reliability Improvements
& Reliability Investments
Linear regression of In (Improvement Ratio) on In (Investment/APUC) gives: ZI APG E3 v1
y =.4718684 Log (Investment/APUC) — 1.0053 ad Ss
10- 5 etzent Confidence Interval = ee
8 me _-7 F100 bot
6 F a -
5 oo
4 2 “ “toaiesse~ * A-10 GPS
3 ie - ee _
aes
275 a
£ “CAT aera aetQ eee
2 “Apache LTU ae i
148 ~opcesin ; e
ost = - "OBIGES (Phedatey” “Apache Gyro
£ : _0isbal Hawk --
0.6 B Lee -
015 | nies BT ‘
0.4 a =
034 a ; 2.419
a = Investment = ini lids = XAPUC
0.3659
T T T T
0.6 0.8 4 2 3 4567849 20 30 405060 80499 200 -300.400.500700 1000 ©2000
Investment/ APUC
Source: Life Cycle Cost Savings by Improving Reliability, Dr. Charles E. McQueary
Director, Operational Test and Evaluation, J anuary 15, 2009
www.qw-itea.org/.../McQuearyGW-ITEAluncheonP resentation] an2009.ppt
As an illustrative data point on the graph of Figure 1, the research determined that the
Predator program invested a total cumulative amount of $39.1 million in reliability
investments over a nine year period. The ratio of this investment to the Average
Production Unit Cost (APUC) of $4.2 million is 9.3 and is the value of the x axis for the
Predator data point. The research also determined that the overall failure rate of the
Predator was reduced by 48.1 percent, resulting in an overall improvement in MTBF
from 40 hours in FY 98 to 77 hours in FY 06, or a 92.5 percent improvement in reliability.
This is the y-axis point for the Predator. The other data points on this graph reflect the
results of similar analysis.
It should be noted that this chart relates reliability investments to reliability
improvements but does not take the next step and relate investments in reliability to
reductions in life-cycle costs. Additional research is focusing on that next step using the
Cost Analysis Strategy Assessment (CASA) model, a total life-cycle cost analysis tool,
and other analytical techniques. (Forbes, Hees, Long, and Stouffer, 2007) Such models
give estimates for changes in twenty year support costs based upon a variety of input
assumptions including changes in reliability. These models, however, do not give
indications of changes in readiness levels or of payback time for the investment.
According to the Department of the Army Economic Analysis Manual, the Break Even
Point (Payback period) is an important metric for investments. For example, two
investments might have similar benefit cost ratios or similar savings to investment
ratios, but if one has a substantial faster payback, it is the superior investment. Both the
readiness and time dynamic aspects of reliability investments need to be included in a
benefits analysis.
One approach to investigating sustainment costs that incorporates both readiness and
time of payback is to view the support process as an on-going enterprise supply chain
that provides new parts, repair, support, maintenance, etc. over the operating life cycle.
Improvements in reliability clearly affect the operational aspects of the supply chain
through reductions in removals, overhaul requirements, new spare acquisitions,
shipment of replacements and all of the associated and related costs. Simulation of this
enterprise using a dynamic modeling approach can establish a relationship between
improvements in reliability and reduced operating costs as well as indicating changes in
readiness levels and time of payback.
Analytical Approach
System Dynamics is a well-suited tool for understanding the structure and dynamics of
complex supply chain systems and the factors over time determining lifecycle costs.
From its very beginning, System Dynamics has been used to analyze supply chains as
a modeling and simulation tool for policy analysis. Forrester’s (1958) groundbreaking
article in the Harvard Business Review demonstrated fundamental supply chain
dynamic behavior such as how small changes in retail sales and promotional activity
can lead to large swings in factory production, i.e., the so called bullwhip or Forrester
effect. Forrester’s model, however, included factory, distribution, and retail tiers in the
supply chain, but no suppliers to the factories. In 2000, J ohn Sterman expanded on
Forrester’s supply chain models, including suppliers linked to the factories. Huang and
Wang (2007) explored the bullwhip effects in a closed loop supply chain system.
Simchi-Levi (2008) and Lee (1997) addressed the bullwhip effects from an analytical
perspective. Schroeter and Spengler (2005) addressed the strategic management of
spare parts in closed loop supply chains. Angerhofer (2000) presented an in-depth
discussion of system dynamics modeling in supply chain management. Killingsworth,
Chavez, and Martin (2008) analyzed the government ordering process within a system
dynamics in two forms: including the extended supply chain and excluding the extended
supply chain. The intent of the current research is to analyze the impacts of
improvements in reliability on supply chain behavior and lifecycle costs using a system
dynamics model. By using appropriate discount and inflation rates, cumulative and
annual costs are measured in relation to investment amounts to weigh the overall
benefits for improvement in the government supply chain.
Model Description
An overview of the supply chain for high-value aviation parts is shown in Figure 2. This
diagram illustrates the flow of parts from new production and overhaul to the final
customer. The overall supply chain process is managed in a feedback form by the
government's ordering or requirements determination process. These algorithms are
typically embedded in a computerized process utilized by item managers, such as the
Army's Supply Control Study. Based upon the calculated recommendations, repair
action or procurement action will be initiated. This process, or something similar, is
used by most government and defense supply chains for high-value parts. (Rosenman,
1964)
Figure 2: Overview of Supply Chain Model
Funding
i
it '
ia f ¥ i
it t Hi !
rt i) H 1
Hes | 1 i 1
BH t 1 Data 1 !
ver | 4 ——— OperationHours =
wat 1 i &Rellablity |
Scrapped | 1 New Spare U 4 | !
atRepa | >) Production [~~~ t ! to
Feclty 1 1] istibution H wi
1 Depot 1 Center for Regional Parts, Removals
b=-+) Qverhad = ----- 1 Serviceable Distribution on
! H Inventory Center Aircraft Scrapped in
i i Feld
Conwnercial
Kee) Overhaul f---~~ s
Retumed for Repair
co INFORMATION/DATA FLOW
—— PARTS/MATERIALS FLOW
—— FUNDING FLOW
By systematically comparing current levels of inventory, due-ins, due-outs, and
historical demand levels, the ordering process determines the recommended buys and
repairs. The demand for a part is driven by the total number of installed parts, monthly
operating hours, and failure rate per part per operating hour, sometimes expressed as a
mean time between failures (MTBF). The supply of parts comes from three possible
sources: production of new items, commercial overhaul of damaged parts, and
government overhaul of damaged parts. Once the production or overhaul process is
completed, parts are then transferred to the central distribution inventory. Each region
has an inventory of key spare parts, and these inventories are replenished from the
central distribution inventory. Parts excessively damaged and unable for repair may be
scrapped at two different points once they are removed from the aircraft. The first
possibility is for parts to be scrapped in the field and not be returned for overhaul. The
second possibility is for parts to be scrapped at the repair facility be it either at a depot
or commercial manufacturer.
Several levels of calculation are incorporated into the supply control ordering process to
determine recommended buys and repairs. (Killingsworth, Chavez, and Martin, 2008)
The determination process calculates the procurement action for new spare parts by
calculating the difference between the procurement reorder point and the total net
assets, and then adding the procurement cycle requirement and the inventory
necessary to meet demands until the next scheduled order. Total net assets are
calculated from due-ins from procurement and repair plus inventories, less due-outs.
The procurement reorder point is based on targeted reserves and safety levels. Within
the model, orders that are placed with the OEM enter production subject to a maximum
production rate and availability of all the required components. Production is completed
after a manufacturing lead time. After production, these new parts flow into the
distribution center for serviceable inventory. Figure 3 illustrates the recommended new
spares procurement action.
Figure 3: Recommended New Spares Procurement Action
Traral (ett 1-1)
at |
i
!
‘ ess
‘ Denne
a INFORMATION/DATA FLOW i % y x
— PARTS/MATERIALS FLOW \ ‘Ar somienents Ronin
. —
FUNDING FLOW pk a
UNDING Peart eptrcenis
one Aca
The determination for recommended repair action differs in an important way from the
determination of the recommended procurement action. First, the maximum
recommended repair action is calculated by subtracting the assets available for repair,
including overhaul and procurement work-in-process less due-outs, from the repair
action point, calculated with reserve levels and safety requirements. This process is
largely driven by historical demands. In the second step, the maximum recommended
repair action, however, is then limited by the unserviceable inventory on hand. A
damaged part must be available for repair or overhaul to take place. The potentially
constrained repair order is allocated between government depot and commercial
overhaul according to capacity levels at each location. The overhaul rates may also be
limited by production capacity levels. Similar to procurement production, once overhaul
is complete, the part is transported to the distribution center for serviceable inventory.
Figure 4 illustrates the process for calculating the recommended repair action.
FIGURE 4: Recommended Repair Action
none INFORMATION/DATA FLOW
—— PARTS/MATERIALS FLOW
FUNDING FLOW
Upon arrival at the distribution center for serviceable inventory, all parts, both new and
repaired, are available for shipment to the regional inventory centers.
The regional inventory center orders parts from the central distribution center to
replenish that inventory being used to replace removed parts. The monthly removals
are dependent upon the number of parts in service (i.e. number installed on aircraft), the
monthly operational hours (i.e. monthly flight hours), and failure rate per monthly
operational hour. Hence, if reliability is improved, the failure rate per flight hour is
reduced, demand is reduced and pull from inventories is reduced. This leads to
reduced orders for new spares and overhaul. This high level view of the model
structure is shown in Figure 5.
FIGURE 5: Reliability, Flight Hours and Removals
Improved
Rellabiliey
é
Scrapped at Repair [scorers ] aoe Falure Rate he aed
Prrafty 9 a ’ a
ry sey ae
Depet and Cantral Ramevenis
»| >| os | Regional | Persson jp Scrapped in
oe aa Imeantory Alrcraft Tedd 56
INFORMATION/DATA FLOW
—_ PARTS/MATERIALS FLOW
The objectives of this research are to determine how investments in reliability
improvement can reduce life-cycle costs and improved readiness. It is assumed in the
model that the investment occurs over a three year period that includes design,
manufacturing, test, and certification. Shorter or longer investment scenarios are easily
included and examined in the model structure. Figure 6 illustrates this structure for the
simulation, and the factors involved in calculating annual current dollar, constant dollar,
and discounted dollar expenditures as well as life-cycle cumulative costs. The figure
also illustrates how the discount and inflation rates are used in the calculation for
constant dollars (Constant year dollars are the result of having the effects of inflation
removed. Constant year dollars are always associated with a base year.), discounted
dollars (Discounted dollars are the present value of a cost made in the future.), and
current dollars (Current year dollars are expressed in the value of the year of in which a
cost is expected to occur, and therefore reflect the effects of inflation).
FIGURE 6: Investment for Improved Reliability
The investment in reliability impacts the spending amounts for each year after the new
improved part is introduced, depending on the degree of reliability improvement for the
part and any changes in the unit cost of the part. It may very well be the case that the
improved part will have a higher production cost and the model enables the
investigation of tradeoff between improved reliability, higher unit cost and reduced
demand. Figure 7 illustrates the calculation in the model of cumulative spending and
annualized spending.
FIGURE 7: Cumulative and Annualized Spending
spending Rate for New Spares
Cost of New Spare New Spare
bepot OVHL
Completion Rate
Repair Cost per Unit
ee ee ae
> Aanualized Spending for
New Spares eo
Cumulative
Spending tar New fe,
Spares —
Completion Rate
Cumulative Spend ra
for OVHL
Commercial OVHL i
Completion Rate -
Annualized Spending
for OVHL
Be Total Annualized
Total Spend for New Spending for New
pares & ovi Spares and OVHI
-- oo?
Evaluation of Alternative Investments
The key objectives of the analysis were twofold:
(i) First, to determine the reductions in life-cycle costs and payback periods for
investments in reliability using “what if" assumptions relating the improved
reliability and the assumed level of investment. In these analyses, payback is
determined and life-cycle costs are calculated assuming for example that a
$10 million dollar investment will generate a 10% improvement in reliability.
The simulation can then be conducted assuming that the $10 million
produces a 20% increase in reliability. These types of analyses allow one to
determine the required reliability improvement of an investment such that an
adequate return is generated through reductions in life-cycle costs. This
enables a business case analysis to be completed for a proposed program.
Second, to determine the reductions in life-cycle costs and payback periods
for investments in reliability using empirical data developed in previous DoD
research that relates investment as a percentage of unit cost to improvements
in reliability. This analysis can then supplement and support a business case
analysis as conducted above.
(ii)
In the first part of the investigation, the model structure was parameterized for a major
repairable helicopter part. The simulation begins 1) anuary 2001. Demands in 2001
and 2002 were fairly stable as were inventories. However, with the onset of the conflict
in Iraq, demands rose sharply in 2003 and have remained elevated. As a result of the
increased demands, inventories of many aviation parts were seriously depleted for a
period of three to four years. For the part in question, the Production Lead Time (PLT)
was roughly 24 months. Because of this long lead time, inventories for this part were
reduced to near zero levels. A key validation test of the model was the ability to capture
these dynamics. In the simulations for exploring improvements in reliability, it is
assumed that investments in reliability were initiated in 2003 and extended through the
end of 2005, a three year investment period. The improved part becomes available at
the beginning of 2006. The time span for this simulation was selected in order to
simulate an initial steady level in demand, then a significant surge in demand and to
investigate the likely impacts of improved reliability on inventory recovery and cost
reductions.
All cases assume stable demand levels from year 2001 to the beginning of 2003. In
addition, all cases include a rise in demand beginning in year 2003 due to an increase
in operating flight hours. In 2003, flight hours increase from 14 hours per aircraft per
month to 18 hours per aircraft per month. Figure 8 presents operating hours per month
over the simulation. It is important to note that this assumption for flight hours is the
major external assumption driving the simulation and that this assumption can be easily
altered.
FIGURE 8 & 9: Operating Hours Per Month Per Aircraft and Monthly Demands
LAT
125 5
10 0
2001 2003 2005 2007 2009 2011 2013 2015 2017 2019 2001 2003 2005 2007 2009 2011 2013 2015 2017 2019
Year Year
Operating Hours Recuring Demands Program Demands
The flight hour assumption and the failure rate per flight hour yields an initial monthly
demand of eleven that increases to fourteen following the increase in flight hours. In
addition to these recurring demands, it is assumed in the model that the part has an
initial monthly program demand of three and that the program demand increases to five
during 2003. These program demands arise through programs such as Reset and
Recapitalization and are independent of flight hours. The demand assumptions are
shown in Figure 9. For the simulation, it is assumed that with growing demands, an
investment program is undertaken to improve reliability.
Note again that for those cases assuming a reliability investment, the investment is
equally divided over three years (2003-2005 in these simulations). Starting in 2006, the
introduction of the part with improved reliability begins to reduce demands, support
inventory levels, and reduce procurement and repair actions. It is assumed in the
model that the new parts are introduced through attrition. That is, as older parts are
removed, they are replaced with the improved part. It is assumed that the removed
parts which are not scrapped are transformed in the overhaul process to the part
configuration with higher reliability. Current research is addressing the case in which
this improvement is not possible in the overhaul process. The complete turnover in parts
10
requires approximately eight years given the demand level in Figure 9. Thus, the
overall transition to the improved MTBF occurs over that period of time.
In some of the alternative cases with improved reliability, an increase in the cost of the
partis assumed. In these cases, the cost increase takes effect in year 2006 as the new
parts are introduced. Table 1 presents the key assumptions for five reliability cases to
be analyzed through simulation.
Evaluation of Investment in Reliability
TABLE 1: Cases for Improved Reliability
Case Reliability Improvement in Parts Cost
Investment Reliability (MTBF) Increase
1 No 0% 0%
2 Yes 33% 0%
3 Yes 33% 15%
4 Yes 50% 0%
5 Yes 50% 15%
Case 1: Base Case Analysis: No investment in reliability, no improvement in reliability,
and no increase in parts cost; This Simulation Should Reflect Actual Supply Chain
Performance;
Case 1 was conducted to provide a base case for inventory levels and procurement and
repair actions over time. Cases 2-5 with improved reliability may then be compared to
this base case. Figures 10 - 12 present the results of Case 1. As seen in Figure 10,
the surge in demand in year 2003 causes a dramatic reduction in serviceable
inventories. This decline in inventories creates backorders in the system, as orders
cannot be completed due to lack of supply in inventory. In Figure 11, new procurement
remains steady for the first four years, but ultimately ramps up due to the increase in
demand arising from the greater number of flight hours. The delay in this ramp-up
arises from the fact that the requirements determination process uses a twenty-four
month rolling average for demand calculations. This rolling average only slowly reflects
the higher demand levels. On Figure 11, the green tick marks represent procurement
orders that arise as the total net assets level drops below the requirements objective. In
Figure 12, the max repair activity rises sharply due to increased demand. The
maximum repair activity represents the repair action that the system would optimally like
to realize, but that action cannot be executed due to the lack of unserviceable inventory.
Even though the max repair action is quite high, only the available unserviceable parts
can be repaired, and the repair action remains at a modest level constrained by the flow
of returning parts for overhaul.
11
Figure 10: Case 1 Inventory Levels
150
wh)
0
2001 2004 2007 2010 2013 2016 2019
Year
Unserviceable Inventory
Setviceable Inventory
Available Inventory at Regions
Backorders
Figure 11: Case 1 Procurement Action Figure 12: Case 1 Repair Action
400 LARP cannainnnan 100
200 50 {iN
| A
UME EEE 0
2001 2004 2007-«-2010 «2013-2016 +2019 2001-2004 2007-2010 -«2013—«2016 «2019
Year Year
Total Net Assets Repair Action
Procurement Reorder Point, Unserviceable Inventory
Procurement A ction, $$ Max Repair Action
Notice that although demands increased sharply in 2003, procurement and repair
actions ramp up slowly. This is again because the typical DoD requirements
determination process uses a twenty four month rolling average as the basis for
demand forecasting. This lagged average introduces a substantial delay in the process
and, when combined with a twenty four month production lead time, creates a situation
where inventories are rapidly pulled down following a sustained surge in demand and
inventory recovery is very slow.
Case 2: Investment made to improve reliability, 33% improvement in reliability, and no
increase in unit part cost.
Case 3: Investment made to improve reliability, 33% improvement in reliability, and a
15% increase in unit part cost.
Cases 2 & 3 examine the impacts of investments in reliability on demands, inventories,
readiness, and life-cycle costs relative to the Base Case. Figures 13 - 15 present the
results of these cases. It should be noted that although Case 3 involves a percent
increase in unit part cost after the investment period, inventory levels, procurement
actions, and repair actions are the same in both cases because it is assumed that
funding is available to make the recommended buys and overhauls. The increase in
cost of the part does, however, impact total life-cycle cost and reduces the return in
12
Case 3. Figure 13 presents simulation results for inventories for Cases 2 and 3. As
may be seen, with the more reliable part entering service in 2006, inventories recover
faster because of the improved reliability and reduced demands. In fact, because of the
long production lead time and demand averaging, orders continue to be made at a
higher than necessary level and inventories overshoot the objective before falling back
to the steady goal appropriate for the new reduced demand levels. The unserviceable
inventory level also rises because fewer parts require repair action due to improved
reliability and reduced demand. Figure 14 shows a reduction in orders for new spares
(the green tick marks are buys). Following the introduction of the improved part, the
interval between orders increases because the parts are now more reliable and monthly
removals are reduced. Lastly, Figure 15 shows a ramp up in unserviceable inventory
after the investment period for Cases 2 and 3. As stated above, the improvement in
reliability causes the unserviceable inventory level to rise due to reduced repair actions
because of longer lasting parts. Thus, unserviceable inventory will build as fewer parts
are overhauled. This may be seen in the lower levels of repair activity.
Figure 13: Cases 2 & 3 Inventory Levels
150
75
0
2001 2004 2007 2010 2013 2016 2019
Year
Unserviceable Inventory
Serviceable Inventory
Available Inventory at Regions
Backorders
Figure 14: Cases 2 & 3 Procurement Action Figure 15: Cases 2 & 3 Repair Action
200 50
set
O MALONE | 0
2001 2004» 2007-2010 «2013'-=«-2016~—«2019 2001 2004 +2007 «2010 :«2013-~«2016 +2019
Year Year
Total Net Asset, £@@— Repair Aion. $£$ —€@@ —£—
Procurement Reorder Point’ HHH Unserviceable Inventory. tH
Procurement ction, $$ Max Repair ction, $$$
Case 4: Investment made to improve reliability, 50% improvement in reliability, and no
increase in unit part cost.
Case 5: Investment made to improve reliability, 50% improvement in reliability, and a
15% increase in unit part cost.
13
Cases 4 and 5 were conducted to assess the impacts of even greater improvements in
reliability. Figures 16, 17 and 18 present the results for Cases 4 and 5. Figure 16
shows that a 50% improvement in reliability greatly increases the sharpness and
rapidity of inventory recovery. Also, the total number of backorders is lower than in any
other cases. Figure 17 shows that in this simulation, once the investment period ends,
the new improved parts are shipped in 2006 and 2007 at close to historical levels.
However, since parts are lasting longer, fewer new parts are required to meet the
requirements objective (RO). As a result, procurement actions are less frequent from
2008 to 2013. They then settle to a lower level that reflects the lower monthly demands.
Similarly, Figure 18 shows that after the investment period, repair actions slow for one
year again because of reduced demand arising from increased reliability in parts and
then repair actions become somewhat less stable as the system adjusts to the new
demand levels. Also, unserviceable inventory rises higher than any of the previous
cases because less repair activity is required.
Figure 16: Cases 4 & 5 Inventory Levels
150
75
0 é
2001 2004 2007 2010 2013 2016 2019
Year
Unserviceable Inventory
Serviceable Inventory
Available Inventory at Regions
Backorders
Figure 17: Cases 4 & 5 Procurement Action Figure 18: Cases 4 & 5 Repair Action
” lta TL LY °
200 50
ret
°
UOT 0
2001 2004 --2007««2010-—«2013-~«2016 +2019 20012004 —-2007—«2010-=«2013—«2016 «2019
Year Year
Total Net A sse’s| $$$ ia iii Repair Action
Procurement Reorder Point. Unserviceable Inventory
Procurement ion, $m i ii i Max Repair Action
Figure 19 presents the current dollars (inflated dollars) annual spend on parts over the
course of the simulation for all five cases. After the investment period, all cases show
significant savings in annual spending relative to the Base Case 1. Case 5 shows an
annual savings of $60 million for this single part in 2020. Figure 20 presents the current
dollars cumulative spend over the ten year period of the simulation. For Case 5, the
cumulative savings are roughly $600 million over this life-cycle assessment. Clearly, the
14
greater the reliability improvement, the greater the total cost savings and these savings
are significant.
FIGURE 19: Current Dollars Annual Spend Cases for Improved Reliability
Current Dollars Annual Spend
$180,000,000
$160,000,000
$140,000,000
w $120,000,000
-& $100,000,000
8 $80,000,000
$60,000,000
$40,000,000
$20,000,000
$0 te
2001 2003 2005 2007 2009 2011 2013 2015 2017 2019
Time
FIGURE 20: Current Dollars Cumulative Spend Cases for Improved Reliability
Current Dollars Cumulative Spend
$2,000,000,000
$1,800,000,000 7
$1,600,000,000
$1,400,000,000 Cases
£ $1,200,000,000 Case 2
3 oat en on =t-Case 3
Q $800,000,000 +
600,000,000 sparCaseit
$400,000,000 = Case 5
$200,000,000
So 4B rr
2001 2003 2005 2007 2009 20112013 201520172019
Time
Figure 21 presents current dollars annual spending amounts for all five cases. For
Cases 2 through 5, the “Savings” columns present the reduction in spending from Case
1 over the twenty year period. The yellow shading denotes the years in which an
investment is made for improved reliability. The table also provides the payback or
break-even (B/E) point in number of years. The lower portion of the chart presents the
15
payback ratio for the investment. This final financial metric is obtained by dividing the
current dollars cumulative savings by the total investment. The payback ratio
demonstrates the time significance of the cost reduction for process improvement. For
example, the results from Case 2 illustrate current dollars cumulative savings of roughly
$75 million in the six years (2006 through the end of 2011) following the investment
period. The chart indicates that a $3 million dollar investment spread equally over years
2003, 2004 and 2005 would be recaptured in 2.27 years after the investment period
ends. A $6 million dollar investment spread equally over those same years would allow
for a payback period of 2.66 years. To evaluate the alternative cases with different
investment amount and payback periods, investment amounts of $3 million, $6 million,
$9 million, and $12 million were used. Similarly, Cases 3, 4, and 5 also exhibit quick
payback or break-even time in years. These results indicate the strong potential for
reduced O&S costs through reliability investments.
FIGURE 21: Annual Spending & Savings for Cases with Improved Reliability
Current Dollar Annual Spending and Savings
Base Scrap/Loss Rate 35%, 15%
(% Improvement in Reliability, % Parts Cost Increase)
* All Cost in Millions
Case 1 Case 2 Case 3 Case 4 Case 5
Year (0%,0%) | (33%,0%) | Savings | (33%,15%) | Savings | (50%,0%) | Savings | (50%,15%) | Savings
2001 $38.14 $38.14 1 $0.00 $38.14 1 $0.00 $38.14 4 $0.00 $38.14 1 $0.00
2002 $40.64 $40.64 | $0.00 $40.64! — $0.00 $40.64! $0.00 $40.64" — $0.00
2006 $81.79 $81.79 | $0.00 $81.79! $0.00 $81.79 § $0.00 $81.79 ! $0.00
2007 $89.49 $88.60, $0.89 $96.28; -$6.79 $88.15 4 $1.34 $95.79 5 -$6.30
2008 $92.00 $84.23 | $7.17 $91.23 § $0.77 $77.36! $14.64 $83.95! $8.05
2009 $89.61 $71.92 | $17.69 $77.37 | $12.24 $26.77 $22.14
2010 $90.79 $68.14 1 $22.65 $72.64 4 $18.15 $33.44 $30.14
2011 $95.14 $67.49 | $27.65 $71.60 | $23.54 $41.81 $39.22
2012 $100.10 $68.23 1 © $31.87 $72.18 1 $27.92 $51.13 1 $48.97 $46.83
2013 $107.26 $72.91 | $34.35 $77.01 | $30.25 $55.26 | $52.00 $49.87
2014 $113.85 $80.41 1 $33.44 $85.18 1 $28.67 $63.11 1 $50.74 $47.83
2015 $120.88 $87.46! $33.42 $93.20! $27.68 $49.03 $44.93
2016 $128.33 $94.38, $33.95 $101.01 5 $27.32 $50.88 $45.72
2017 $136.29 | $101.83! $34.46 $109.16! $27.13 1 $52.06 $46.14
2018 $144.73 | $108.92, $35.81 $116.83; $27.90 $90.47 , $54.26 $47.80
2019 $153.65 $115.94 1 $37.71 $124.37 | $29.28 $96.44 1 $57.21 $50.29
2020 $163.15 $123.11 | $40.04 $132.07 , $31.08 $102.48 , $60.67 $53.32
Cumulative Savings $391.70 $305.14 $593.82 $525.98
$3 Million | B/E (years) 2.27 3.74 2.11 3.06
$6 Million | B/E (years) 2.66 3.98 2.32 3.19
$9 Million | B/E (years) 3.02 4.15 2.52 3.33
$12 Million | B/E (years) 3.19 4.32 2.23 3.46
Benefit/investment Ratio
$3 Million 129.57 100.71 196.94 174.33
$6 Million 64.28 49.86 97.97 86.66
$9 Million 42.52 32.90 64.98 57.44
$12 Million 31.64 24.43 48.49 42.83
16
Evaluation of Reliability Investment from Empirical Data
In the second part of the investigation, the model structure was again parameterized for
a major repairable helicopter part but rather than for a variable demand rate, a twenty
year steady state life-cycle was assumed. Moreover, this part of the analysis utilizes an
empirical relationship between reliability investment and reliability improvement.
As discussed in the Introduction, LMI with the sponsorship of DoD developed a linear
regression equation relating investments to reliability improvements. This linear
regression is presented in Figure 1. The regression relates the Investment divided by
the Average Production Unit Cost (APUC) to the Reliability Improvement Ratio, which is
the percentage increase in reliability. By utilizing this linear regression, the likely
improvement in reliability arising from a certain investment can be determined. It should
be noted that in Figure 1 the cases in the lower range of the investment ratio tend to be
for large systems or aircraft. As a result, ratios of 1 to 10 may not be appropriate for
major repairable parts. Rather, ratios ranging from 20 to 1,000 may be more
appropriate for investments to improve the reliability of major assemblies and parts. Itis
this upper range of investment ratios that is used in this study.
Table 2 presents four cases developed using the empirical regression between
investment and reliability improvement.
TABLE 2: Cases for Improved Reliability Using Empirical Data
Reduction in
Case ee a Failure Rate
Per Flight Hour
6 0 0 0.0%
7 20 150% 60.0%
8 30 200% 66.7%
9 40 225% 69.2%
For this part of the analysis, Case 6 represents the Base Case before any investment in
reliability improvement is made. It represents “business as usual.” The
“Investment/APUC” column gives for each case the ratio for investment to average
production unit cost (APUC). For example, in Case 7 with an APUC of $250,000, an
investment of $5 million yields an investment ratio of 20. For Case 8, an investment of
$7.5 Million yields a ratio of 30 and for Case 9 an investment of $10 Million yields a ratio
of 40. In the simulation analyses, the total investment amount is divided equally over
three years (Years 1 through 3 of the simulation). The “% Increase in MTBF” column is
calculated from the Linear Regression Example provided in Figure 1. The final column
provides the reduction in failure rate per flight hour from the base failure rate. The
Failures per Flight Hour is inversely related to the MTBF.
Figures 22 - 24 present the inventory levels, procurement actions, and repair actions
over the course of the twenty year simulation for Case 6, the twenty year Base Case.
17
With constant demands, all inventory levels, and procurement and repair actions remain
constant over the simulation as would be expected.
FIGURE 22: Case 6 Inventory Levels
100
50
0
0 24 48 72 96 120 144 168 192 216 240
Y ear (Months)
Serviceable Inventory
Available Inventory at Regions
FIGURE 23: Case 6 Procurement Action FIGURE 24: Case 6 Repair Action
400 7)
300 fas ae AARAAAADABABASSY 60
200 40
100 2»
o LULU TEE PEE o
0 4 48 7 9% «120 M4 168 192216240 0% a8 7296206191 Dae
Time (Month) Tine (Mont)
Procurement Acton Repair Acton
Total Net Assets Unsere Eentory
Procurement Reonier Po << Max Repair Acton
Figures 25 - 27 present the results for Case 7 with reliability improvement of 150%.
Note that these cases include large improvements in reliability. For Case 7, the MTBF
is increased by 150%. This results in monthly removals dropping by more than half, in
fact, a reduction of 60% over the introduction period of roughly eight years. As seen in
Figure 25, inventory levels climb after the investment period (Years 1 to 3) and the new
part begins introduction. The growth in inventory is a result of parts arriving out of a two
year production pipeline into an environment of reduced demand. The model
incorporates and simulates the actions of the requirements determination process which
uses a twenty four month rolling average for demand. As a result, recommended
procurement and repair actions do not begin to slow down in the simulation until over a
year of reduced demands. Figure 26 presents procurement actions. Importantly, six
years after the introduction of the new improved part, new procurement halts for three
years so as to work off the accumulated inventory. Even as new procurement action
begins to pick up, fewer numbers of new parts are ordered than initial levels. Figure 27
shows the slow reduction in repair action and the buildup of the unserviceable inventory.
18
This growth arises because fewer parts are inducted into the overhaul process due to
longer lasting parts and reduced demands.
FIGURE 25: Case 7 Inventory Levels
100
50
COT re Pee
0
0 24 48 72 96 120 144 168 192 216 240
Year (Months)
Serviceable Inventory
Available Inventory at Regions
FIGURE 26: Case 7 Procurement Action FIGURE 27: Case 7 Repair Action
400 80
smn bopnannsn, 0
= [Pr “ ana
100 20
HTT Litter o ae
ua «85 TO ida 192216 240 ar eT
‘Time (Monih) ‘Time (Month)
Procurement A in, $$_$_$__$___ Repair Action
Total Net Assets Unserviceable Inventory
Procurement Reorder Poin, €£$ €£$ Max Repair Action
Figures 28 - 30 present the results for Case 8, assuming a 200% improvement in
reliability. Figure 28 illustrates the inventory levels, which, similarly to Case 7, become
significantly higher following the investment period. Again, this is the result of reduced
demands and historical orders emerging from the production pipeline. As parts become
more reliable and last longer, serviceable inventories, unserviceable inventories, and
available inventories at the regional facilities increase substantially. Figure 29 presents
the procurement action over time. The orders for new parts halt for five years as the
demand is consistently met through inventories and overhaul. Lastly, as shown in
Figure 30, repair orders are reduced over time and that it requires a number of years to
reduce the buildup of unserviceable inventory.
19
FIGURE 28: Case 8 Inventory Levels
100
50
0 24 48 72 96 120 144 168 192 216 240
Y ear (Months)
Serviceable Inventory
Available Inventory at Regions
FIGURE 29: Case 8 Procurement Action FIGURE 30: Case 8 Repair Action
400 80
300 AISA
An AL | eT
nee
200 TAL re hl IN
f|
/ \
100 2 A \
Fea LN
o LETT EL tt piseieigiaes 4 : =
a a a |
Ce Ca =
Procurement Acton ee
Total Net Assets ‘Max Repair Acton
Procurement Reorder Point, $$$
Figures 31 - 33 present the results for Case 9 with a 225% improvement in reliability.
Results are similar except that the buildup of inventories is greater and procurement is
halted for a more extended period of time.
FIGURE 31: Case 9 Inventory Levels
PN
0 24 48 72 96 120 144 168 192 216 240
Y ear (Months)
100
50
0
Serviceable Inventory
Available Inventory at Regions
20
FIGURE 32: Case 9 Procurement Action FIGURE 33: Case 9 Repair Action
400 80
o UATE | I peered
0 2 48 72 96 120 14 ios 192 216 240 0 2 48 «72)~«9%G6 «(20144168 «192 216 240
‘Time (Month) Time (Month)
Procurement etn, $$_$_$_$__ Repair Action
Total Net Assets Unserviceable Inventory
Proarenent Reade? oln=—$£$&@u™i@Oouo™°| Max Repair Action
Figure 34 presents the annual spend in current dollars for the four cases. The
investment expenditures during years one through three are included in the spend for
Cases 7, 8 and 9. Note that the spend for the improved reliability cases drops
significantly in the years following the introduction of the new part. This drop is
associated with the sharp reductions in new procurement that occurs several years
following introduction. When procurement begins to be required again, the spend
begins a slow climb. In all cases however, the annual spend is roughly $60 million
lower than the base cost in Case 6.
FIGURE 34: Comparing Current Dollar Annual Spend for Various Cases
$120.0
$100.0
$80.0
Current Gace
Dollars $60.0 =
(Millions) —liCase 7
$40.0 —t—Case 8
——Case 9
$20.0 -
S- LS a 4
12345 67 8 9 1011121314151617181920
Time (Year)
Figure 35 presents cumulative spend in current dollars for the four cases. The
reduction in spending from the base over the twenty year life cycle ranges from $700
million to $800 million, thus, demonstrating the substantial returns provided by the
investments that ranged from $5 million to $10 million for a part costing $250,000.
21
FIGURE 35: Comparing Current Dollar Cumulative Spend for Various Cases
$1,600.0
$1,400.0
$1,200.0
$1,000.0
Current $800.0
Dollars .
(Millions) $600.0
$400.0 +
$200.0
$-
12345 67 8 9 1011121314151617181920
Time (Year)
—e—Case 6
—HCase7
—i=Case 8
——Case9
Table 3 and Figure 36 present a cumulative lifecycle summary for Cases 6, 7, 8, and 9.
As may be seen, paybacks from investment in reliability can be very substantial and
very attractive. Note that the investment amount is included in the costs of the
alternatives. As Figure 36 illustrates, there is a point where the percentage in cost
reduction begins to level off and decline. Savings reach an upper limit of approximately
50% of base costs and then decline as the investment amount increases and ultimately
increases the total costs. Nevertheless, for the part with APUC of $250,000,
investments in improved reliability on the order of $7.5 to $10 million generate estimated
life cycle cost reductions of roughly $600 million in current dollars; this may be
interpreted approximately as needing to buy 1,300 fewer parts over the 20 year life
cycle.
Table 3: Reductions in Life Cycle Costs
Case | Investment tnvesnnenw ee pees pal or Savings pauings! t
%* (Current $)
6 $ ) 0% $ 1,398,720,000 | $ : 0
7 $ 5,000,000 20 150% $ 707,848,000 | $ 690,872,000 49.39%
8 $ 7,500,000 30 200% $ 632,752,000 | $ 765,968,000 54.76%
9 $ 10,000,000 40 225% $ 605,767,000 | $ 792,953,000 56.69%
22
As may be seen in Figure 36, there appears to be an investment “sweet spot” that exists
in a range of (Inv/AP UC) ratio between fifty and one hundred.
FIGURE 36: Constant Dollar Investment/APUC vs. Percent Reduction in Costs
60.0%
50.0% ~~
40.0%
Percent i,
Reduction 30.0%
in Costs
20.0%
10.0%
0.0% ; T r : 1
O 200 400 600 800 1000
Investment/APUC
Figure 37 presents the Benefits (Savings) to Investment ratio as a function of the
investment ratio in constant dollars. This reinforces the finding that an investment ratio
exceeding 100 produces sharply lower returns.
FIGURE 37: Constant Dollar Investment/APUC vs. Return on Investment
30.0
25.0 \
20.0
Benefit/Inv 15.0
Ratio
10.0 NS
5.0
0.0 T r 1
Investment/APUC
23
Impacts on Readiness and Availability
Although the discussion so far has focused on economics, namely investments, costs,
and savings, improvements in reliability can also have significant impacts on aircraft
availability and readiness. Table 4 presents the impacts of improved reliability and
failure rate reductions on aircraft availability and readiness. Failure rate reductions
lower the average monthly removals and lead to increased annual aircraft availability
hours. In this analysis, itis assumed a part in repair would ground a helicopter for three
days (72 hours). The table presents the annual additional hours of availability that arise
from the improvements in reliability assumed in Cases 6 - 9. As may be seen, the
investments not only yield large savings in expenditures, but also provide thousands of
additional hours of availability.
Table 4: Impacts on Aircraft Availability and Readiness
% in Failure ANA
Rate Reduction} Average Unavailable | Unavailable |Annual Reduction Additional
Case | (Failure Rate Monthly Hours per Hours in Aircraft ‘Available
per Flight Demands Year* Reduction % Impacted H
jours
Hour)
6 - 14.0 12,096 : -
7 60.0% 74 6,394 47.1% 79 5,702
8 66.7% 6.7 5,757 52.4% 88 6,339
9 69.2% 6.4 5,519 54.4% 91 6,577
Sensitivity Analysis
A sensitivity analysis was performed to examine the impacts of monthly demand rates
and average production unit cost (APUC) on savings and return on investment. In the
sensitivity analysis, all cases assume a 150% improvement in reliability. From the
empirical relationship in Figure 1, this means an Investment to APUC ratio of twenty.
Thus for a part costing $250,000, the investment required is $5,000,000. For the part
costing $500,000, the investment required is $10,000,000. As may be seen in Figure
38, the payback period in years (breakeven point) is achieved more quickly for a part
with higher demands because of the greater savings for high volume parts. In addition,
more expensive parts also exhibit faster paybacks than less expensive parts, again
because of the greater savings. Figure 39 illustrates the cumulative savings for cases
with higher demand rates and more expensive unit costs. Again, the greatest amount of
cumulative savings is gained through helicopter parts with higher monthly demand and
higher unit cost, as both variables directly impact cumulative savings over a system's
useful life.
24
FIGURE 38: Payback Period with Varying Monthly Demands and Unit Costs
Payback
Period (Break
Even Point in
Years)
3.00
2.00
1.00
0.00
APUC
=m $250,000
—t= $350,000
—<$500,000
14 18 22 26
Monthly Demands
FIGURE 39: Cumulative Savings with Varying Demands and Unit Costs
Current
Dollar
Cumulative
Savings
($Billions)
as
al APUC
== $250,000
—@ $350,000
te $500,000
14 18 22 26
Monthly Demands
25
Conclusions
“Doing more with less” has become a long-running and recurring theme across the
globe. This challenge is especially acute for the US Department of Defense and the
branches of the armed services where demands are great and budgets are tight.
Because typically the costs to operate, maintain, and dispose of a weapon system
account for about 72 percent of the total cost of ownership, much effort has focused on
these expenditures. Long-term costs incurred in maintenance are subject to increased
focus. Affordability is now a key metric. Improvements in reliability have multiple cost
saving impacts such as fewer parts to buy and overhaul, smaller inventories of
replacement parts, fewer inspections, reduced maintenance hours and down time,
reduced transportation costs to ship replacement parts, and others. Because of the
complexities of the supply chain and procurement processes, financial evaluation of
reliability investment is very difficult. The requirements determination process, long
production lead times, and target inventories must also be considered. A system
dynamics model has been developed of an aviation supply chain that enables
evaluation of alternative cases in which investments are made to improve reliability,
lower total demands, and reduce spending on new procurement and overhaul over the
lifecycle. It is shown that the payback potential of an investment depends upon annual
demand for the part, cost of the part, percent improvement in reliability achieved, and
any increase in cost of the part due to the re-design. The analysis show that returns
can be high and payback periods can be fast, particularly for investments to improve
reliability of items with high demand and high cost. Moreover, the simulation results
indicate an investment “sweet spot” may exist in a range for the ratio of
Investment/P roduct Cost between fifty and one hundred. In this range, the percentage
reduction in cost is maximized. The research also indicates that close coordination is
needed between program management, procurement planning and acquisition in order
to fully realize savings. Ongoing research is developing reliability investment strategies
and estimates for lifecycle costs under differing demand, manufacturing and overhaul
scenarios.
26
References
Abramsom, Mark A., and Roland S. Harris Ill, ed. The Procurement Revolution. New
York: Rowman and Littlefield Publishers, Inc., 2003.
Aerospace Industries Association. “Life Cycle Sustainment: Accomplishing
Performance Driven Outcomes through Condition Based Maintenance and Continuous
Process Improvement.”
Ain, Aron. “Doing More With Less”. Business Week. April 22, 2009.
Angerhofer, Bernard J., and Marios C. Angelides. “System Dynamics Modeling in
Supply Chain Management Research Review.” Paper presented at the 2000 Winter
Simulation Conference, Orlando, FL, December 10-13, 2000.
Anthony, Scott. “Creative Innovation: Doing More With Less.” Forbes. February 26,
2009.
Correa, Edilia. “Aviation Engineering Reliability and Sustainment Programs.” Defense
Logistics Agency. Defense Supply Center Richmond, October 31, 2007.
Department of Defense. Reliability, Availability, Maintainability, and Cost Rationale
Report Manual. Presented by the Office of the Secretary of Defense in Collaboration
with The J oint Staff, J une 1, 2009.
Forbes, James, Jing Hees, Andrew E. Long, and Virginia Stouffer. “Empirical
Relationships Between Reliability Investments and Life-Cycle Support Costs.” LMI
Government Consulting. Report SA701TI. J une 2007.
Forrester, J ay W. Industrial Dynamics. Cambridge: MIT Press, 1961.
Forrester, Jay W. “Industrial Dynamics: A Major Breakthrough for Decision Makers.”
Harvard Business Review (J uly-August 1958).
Gansler, Jacques S., and Robert E. Luby Jr. Transforming Government Supply Chain
Management. New York: Rowman and Littlefield Publishers, Inc., 2004.
GAO. “Army Inventory: Army Annually Spends Millions to Keep Retention-Level
Stocks.” NSIAD-90-236, September 11, 1990.
GAO. “Army Inventory: Parts Shortages are Impacting Operations and Maintenance
Effectiveness.” GAO-01-772, J uly 2001.
GAO. ‘The Army Should Improve Its Requirements Determination System.” PLRD-82-
19, December 1, 1981.
GAO. “GAO: Inventory Management: DOD Can Build on Progress by Using Best
Practices for Reparable Parts.” GAO-NSIAD-98-97, February 1998.
GAO. “Defense Logistics: Improved analysis and Cost Data Needed to Evaluate the
Cost-Effectiveness of Performance Based Logistics.” GAO-09-41, December 2008.
GAO. “Defense Inventory: Opportunities Exist to Improve the Management of DOD's
Acquisition Lead Times for Spare Parts.” GAO-07-281, March 2, 2007.
27
GAO. “DOD's High-Risk Areas: Efforts to Improve Supply Chain can be Enhanced by
Linkage to Outcomes, Progress in Transforming Business Operations, and
Reexamination of Logistics Governance and Strategy.” GAO-07-1064T, J uly 10, 2007.
GAO. “Setting Requirements Differently Could Reduce Weapon Systems’ Total
Ownership Costs.” GAO-03-57. February 2003.
Gottlieb, Hildy. “Doing More With Less in Hard Times.” The Chronicle of Philanthropy.
December 9, 2008.
Huang, Lizhen, and Qifan Wang. ‘The Bullwhip Effect in the Closed Loop Supply
Chain.” Paper presented at the 2007 International Conference of the System Dynamics
Society and 50" Anniversary Celebration, Boston, MA, J uly 29 - August 2, 2007.
Killingsworth, William R., Regina K. Chavez, and Nelson T. Martin. “The Dynamics of
Multi-Channel Supply Chains for High-Value Government Aviation Parts.” Paper
submitted to the 26" International Conference of the System Dynamics Society, Athens,
Greece, J uly 20-24, 2008.
Killingsworth, William R., Regina K. Chavez, and Nelson T. Martin. “The Dynamics of
the Government Supply Process for High-Value Spare Parts.” Paper submitted to the
26" International Conference of the System Dynamics Society, Athens, Greece, J uly
20-24, 2008.
Kim, Sang-Hyun, Morris A. Cohen, and Serguei Netessine. “Reliability or Inventory?
Analysis of Product Support Contracts in the Defense Industry.” September 10, 2009.
Lee, Hau L., V. Padmanabhan, and Seungjin Whang. ‘The Bullwhip Effect in Supply
Chains.” Sloan Management Review 3837 (Spring 1997): 93-102.
McQueary, Charles E. “Life Cycle Costs Savings by Improving Reliability.” Operational
Test and Evaluation, Office of the Secretary of Defense, J anuary 15, 2009.
Minnich, Dennis, and Frank Maier. “Supply Chain Responsiveness and Efficiency -
Complementing or Contradicting Each Other?” Paper presented at the 2006
International Conference of the System Dynamics Society, Nijmegen, Netherlands, J uly
23-27, 2006.
Minnich, Dennis, and Frank Maier. “Responsiveness and Efficiency of Pull-Based and
Push-Based Planning Systems in the High-Tech Electronics Industry.” Paper presented
at the 2007 International Conference of the System Dynamics Society and 50"
Anniversary Celebration, Boston, MA, J uly 29 - August 2, 2007.
Rosenman, B. Hoekstra, D. “A Management System for High-Value Army Aviation
Components.” IRO Report, October 1964.
Rosenman, Bernard B. “CCSS Supply Management.” IRO Report No. 280, November
1980.
Rosenman, Bernard B. “Supply Control Study Instability.” IRO Report No. 285, J une
1981.
28
Shroeter, Marcus, and Thomas Spengler. “A System Dynamics Model for Strategic
Management of Spare Parts in Closed-Loop Supply Chains.” Paper presented at the
23" International Conference of the System Dynamics Society, Boston, MA. J uly 12 -
21, 2005.
Shute, Toby. “Energy Companies Doing More With Less”. Motley Fool. August 5,
2009.
Simchi-Levi, David, Philip Kaminsky, and Edith Simchi-Levi. Designing and Managing
the Supply Chain: Concepts, Strategies and Case Studies. New York: McGraw-Hill
Irwin, 2008.
Sterman, J ohn D. “Business Dynamics: System Thinking and Modeling for a Complex
World.” McGraw-Hill Companies, Inc., 2000.
Teresko, J ohn. “The Boeing 787: A Matter of Materials.” Industry Week, Dec 2007.
Thorne, Steve C. “Rightsizing DOD Inventory: A Critical Look at Excesses, Incentives
and Cultural Change.” Master’s Thesis, Naval Postgraduate School, 1999.
29
