General Disclaimer One or More of The Following Statements May Affect This Document
General Disclaimer One or More of The Following Statements May Affect This Document
This document has been reproduced from the best copy furnished by the
organizational source. It is being released in the interest of making available as
much information as possible.
This document may contain data, which exceeds the sheet parameters. It was
furnished in this condition by the organizational source and is the best copy
available.
This document may contain tone-on-tone or color graphs, charts and/or pictures,
which have been reproduced in black and white.
Portions of this document are not fully legible due to the historical nature of some
of the material. However, it is the best reproduction available from the original
submission.
Bryan DeHoff4
SynGenics Corp. Delaware, Ohio, 43015
Daniel J. H. Levack2
Pratt & Whitney Rocketdyne, Canoga Park, Calfornia, 91309
and
Russel E. Rhodes3
NASA, Kennedy Space Center, Florida, 32899
'Principal Aerospace Engineer, SynGenics Corp, 5190 Olentangy, River Road, Delaware, OH 43015, AIAA Senior
Member.
2 Program Manager, Advanced Programs, P.O. Box 7922 / MS RFB19, A1[AA Member.
AST, Technical Management, Engineering Directorate Design & Development Eng Div Sys Engineering &
Integration Br, Kennedy Space Center, FloridalNE-D2, AIAA Senior Member..
Nomenclature
CONSIZ = Configuration Sizing (A LaRC vehicle sizing code)
DDT&E = Design, Development, Test and Evaluation
DoD = Department of Defense
ESR&T = Exploration System Research and Technology
ETO = Earth-to-Orbit
FBS = Functional Breakdown Structure
HDBK = Handbook
KSC = NASA Kennedy Space Center
Li = Earth-Moon Lagrange point
LaRC = NASA Langley Research Center
LCC = Life Cycle Cost
MIL STD = Military Standard
NASA = National Aeronautics and Space Administration
NGLT = Next-Generation Launch Technologies
SPST = Space Propulsion Synergy Team
WBS = Work Breakdown Structure
I. Introduction
HE evolution of rockets is an evolution of complexity, and increasingly complex systems present many
T (squared) opportunities for errors, omissions, or bad assumptions. Decomposing complex systems is the science
of Systems Engineering; a science which matured alongside the complex systems of the Apollo era. However, like
most true scientific knowledge, the origins of the science of Systems Engineering can really be found in the thinking
and actions of those early engineers and logisticians from the 1600's era of exploration. As they studied the
problems before them, the successes in exploration added continents to the map, identified the currents in the ocean,
and solved the problems of navigation. Today, the field of Systems Engineering allows the engineer to deal with
increasing complexity using a disciplined analysis of the problem to be solved or the goal to be achieved.
The problem of this paper is the problem of exploration, which has similar aspects if one is venturing from
Europe across the seas in the 1600's, exploring the polar regions in the late 1800's or exploring space in the here and
now. Each of these exploration ages has dealt with the same system issues - how to predict the equipment and
supply requirements for the unforeseen experiences that will be encountered by the explorers. Civilian and military
applications of space transportation have been pursued for 50 years and there has been, and still is, a need for safe,
dependable and affordable space transportation systems, both in the regimes of Earth-to-Orbit (ETO) and in-space.
Safety, dependability, and affordability must be met to enable sustainable space exploration or conunercial space
development to occur. In the recent past, several approaches which focused on performance while trying to satisfy
these needs have been explored. Fully expendable and partially reusable space transportation systems have been
developed and put in operation. However, all current space transportations systems have major shortfalls. More
importantly, the design process tools have major shortfalls for focusing on the goals of safety, dependability, and
affordability.
flight systems weight. The present cost estimating process bases cost on flight system dry weight and performs all
trade studies optimizing each single flight system function, with no effort to address integrating the total systems
considering all desired attributes. As an example, objectives were set for LCC for the Shuttle, but no engineering
management processes were exercised to provide control (only the DDT&E cost was tracked). Because the objective
was not treated as a requirement, but only as a goal, it was not achieved.
The SaturnlApollo lunar exploration program was terminated early because the recurring transportation cost was
not sustainable while supporting the exploration efforts. The reusable Shuttle transportation system was developed
to replace the Saturn launch vehicle in an effort to greatly reduce the recurring cost of transportation. Even though
the recurring cost of space transportation systems operation was reduced approximately 75 to 80 percent', the
reduction was not sufficient and did not approach the target goal.
A major lesson identified from the Saturn/Apollo and Shuttle experience is that much improved, innovative
processes must be developed and rigorously applied to effectively control LCC. Any future space transportation
system LCC must be controlled throughout the entire design concept phase, DDT&E phase, and the operations
phase to provide a sustainable space exploration program. Since a major part of LCC for a space transportation
system is the recurring or operational phase cost, this cost must be defined in the design concept phase and
rigorously controlled throughout that and subsequent phases.
The primary lesson learned from studying the past programs is that if the industry always does what it has done
before, then the industry will always get the same result. Therefore, achievement of major reductions in LCC will
not be accomplished using the same existing processes and approaches.
Although many specific problems have been identified in the approaches and tools used in the past to achieve
affordable LCC, this paper will concentrate only on two past problems:
1. architecture options have not been compared on the same level in early concept studies resulting in the
inability to choose architectures based on believable LCC comparisons, and not all needed mission
functions have been fully identified at program initiation;
2. there has been much less functional integration across elements than would be desirable for lowering LCC.
Part of the problem is the lack of good approaches and good tools to identify and assess all of the needed
functions of a specific mission.
the suppliers may require a heat exchanger of some sort. That heat exchanger will require a working fluid. Each of
the suppliers will optimize its heat exchanger to best meet its own mass, performance, non-recurring, and recurring
cost requirements (allocations). Unfortunately, since the internal environments of each supplier are probably
different, the optimum working fluid choice will probably be different. If four working fluids are chosen (or even
very different conditions of the same fluid, e.g., different types of purified water), then both the non-recurring and
the recurring costs incurred by Operations will increase by about a factor of four in the area of building and
maintaining the facilities and supplying the heat exchanger fluids.
The point is that unless Level X uses its information flow from all the suppliers and operations to modify the
choices of the suppliers, the resulting system may be far from meeting its recurring, and maybe even non-recurring,
cost requirements, even though the system does meet its mass and performance requirements and each of the
suppliers met their non-recurring and recurring cost requirements. If costs are requirements and not goals, and thus
enforceable and tradable amongst the other requirements, then Level X can seek the optimum system choice
regarding number of heat exchanger working fluids. The answer may be one or more depending on the degree of
sub-optimal design needed to use specific fluids and the relative impact on the mass, performance, and cost of the
choices.
To make this approach work, Level X must have the ability and will to modify the requirements to each supplier
and operations in an expeditious manner to quickly reflect the system level trades. Thus, Level X must have the
analytic tools, such as utility analysis, to trade between dissimilar requirements; the contract mechanisms to quickly
modify requirements; and the budget to flow-down for re-design as necessary.
The current approach to the development of systems lacks integration to achieve the minimum total hardware..
This occurs because the approach is focused only on each flight element without regard to the impacts on support
infrastructure. There is a need for a structured engineering guide that provides the functional definition necessary at
all levels to allow integration at all levels such that the architecture design can achieve the desired support
infrastructure and labor costs consistent with the overall affordability requirements. There is also a need to provide
visibility of duplication of functions within the overall architecture, across all elements, so that better integration can
be achieved and the goal of significantly reduced LCC can be achieved.
through the entire systems acquisition process, some more than once. The SPST concluded its first task should use
its member's diversified expertise toward developing new "Management Decision Making Tools" or innovative
approaches and processes in the architectural design, development, and operations of space transportation systems to
satisfy the challenging requirements of the transportation operators and the payload customers while providing a
bridge to commercializing space travel.
The SPST has developed a new approach for formulating requirements that will provide full accountability of all
functions required to perform the planned missions. The approach is to develop a top-level functional breakdown
structure (FBS) with modular subsets that may be utilized as a basis for defining the desired "functional
requirements" in any system.
The FBS developed by the SPST evolved out of efforts started at NASA. In an effort to determine where in the
design sequence opportunities are being lost to perform early systems integration, the analysis team reviewed typical
weight statements used by such popular weights and sizing programs as CONSIZ. The team found relatively
detailed functional breakdowns for engine systems and primary structures. Not found was an equivalent level of
definition and identification for the other design disciplines, such as thermal management, communications,
guidance and flight control, and power management. While much of this functional hardware may have been
"covered" at higher indentured levels in the weight breakdown structures, the generic design functions are not being
treated homogeneously during the design process for attributes other than weight; such as total subsystem count,
relative order of magnitude of ground interfaces, number of separate working fluids, propellants and gases, and so
forth. Thus, the conceptual design process is likely to be left blind to these key design characteristics. Further, in
some weight breakdown structures being used, functional hardware required for the integration of flight elements
and stages was going unidentified (such as ordnance subsystems, attach hardware, and thermal protection). For the
operations specialist participating in concept reviews or for the operations modeler, these non-homogeneously
defined system weight breakdown structures were the best available.
The question was raised as to whether one could generically define vehicle and flight element design functions
homogeneously across all the design disciplines, similar to the way the ground functions were generically defined.
This might provide a pathway to reform traditional weight statements into more uniformly defined system
statements that would inherently possess more conceptual-level definition and insight than just mass, volume, and
center of gravity. A cataloging of generic ground functions had already been created and refined several times (see
Section 3 and the FBS in Appendix A of Reference 2). It seemed reasonable to perform a similar, start-from-scratch
exercise for generic flight system definition.
Thus, a Functional Breakdown Structure (FBS) effort was organized in the Systems Engineering Office at KSC
in 2003 in support of NASA's Next-Generation Launch Technologies (NGLT) program. By 2004, the government,
industry, and academia Space Propulsion Synergy Team began actual development work on a FBS. During the
formative stages of the work, the FBS was significantly broadened in scope to support advanced technology life
cycle analysis for NASA's Exploration Systems Research and Technology (ESR&T) effort as part of the Vision for
Space Exploration.
The SPST Functional Breakdown Structure (FBS) has evolved from over 10 years of analyzing the functional
requirements of space transportation systems, including Earth-to-Orbit, and in-space oriented systems. Based on that
experience, the SPST began to see a repeatable pattern of system developments which excluded major requirements
in the early planning, for a variety of reasons. The result of excluding requirements in the early planning, and
attempting to address them later in the system design program has always been significant growth in total life cycle
or system acquisition cost, decreases in system performance, increased development schedule, or all three. Previous
products of the SPST efforts include the "A Guide for the Design of Highly Reusable Space Transportation 3, the
"SPST Collaborative Prioritization of Advanced RLV Technologies" 4, and the Space Shuttle Shortfalls
Assessment5.
the things a space exploration system or space transportation system would have to be capable of doing. The FBS
emphasizes the "what's" that an architecture must do, while the WBS emphasizes the "how's" that the architecture
uses to achieve the "what's". By approaching the systems engineering problem from the functional view, instead of
the element or hardware view, the SPST has created an exhaustive list of potential functions that the various
architecture designers can use to evaluate the completeness of their design concepts.
If one considers the typical aspects of a systems engineering process, such as the "Vee" diagram of Buede 6, the
systems engineering process is only halfway complete if it doesn't address ways to verify and test the design's
ability to achieve the requirements. If affordability or life cycle cost targets are treated as requirements, then the
Systems Engineering process needs to include verification of these requirements as well. By distilling the basic
problem down into the individual capabilities required, architectural details are not brought into the debate until the
appropriate time. Examples abound in everyday life of architectural differences in accomplishing the same function
as one trip through the parking lot will demonstrate. Every five passenger car in the lot meets the same basic
functional requirements, with greatly varying levels of Life Cycle Cost. When the U.S. astronauts first saw the
Soyuz hardware and launch structures during the Apollo/Soyuz Joint Program in the early 1970's, they commented
on the differences of the hardware between the two systems; "The Russians entered the project with equipment
decidedly backward compared with the American hardware and systems."7
Because a traditional WBS is an indentured list of systems/sub-systems/parts, by its nature it needs to reflect a
technical solution to reach deep levels of indenture. Consequently, WBS's are necessary and ideal for weight
management and control and performance comparisons, managing contractors and sub-contractors, and some types
of non-recurring cost comparisons.
On the other hand, because a FBS is an indentured listing of the system's functions for an architecture from the
very top level down to where the last function identified cannot be further decomposed, it is especially useful for
concept definition and trade studies. The FBS is ideal for requirements defrnition because it isolates the functions
from the technical solutions/subsystems/parts. The FBS allows imposition of true functional requirements versus
using constraints due to particular technical solutions, it helps to keep requirements at the same level throughout the
architecture. The FBS is ideal as a filter for concept comparison because it facilitates the use of two criteria: it
checks concepts a, b, c, etc. against the requirements defined at given function levels; and it checks-off functions
implemented by the concept. Fair LCC comparisons require that concepts only be compared when they satisfy both
of these criteria. Concepts that do not satisfy both of these criteria are leaving out some systems/subsystems/parts.
¶Office
_ j 1LL.I-I J J-l)I ri
-Ir1
'!J e Dii ic tnsert Format Tools Data W,dow Hptp JJ-j
- B! U ,
° 55
FH Tenenc Funcnon Descrrprron FBS5rhitrh
:1
• 10 Sujstern Architectural Concept Narno
• 11 Vehicle Element )e 9. Boosrem. Orbiter, Pagload element, repeat as needed For elermrwmtrsj
• I II AirFrame Structure Mechanisms
I 2 Propulsion
lii Power Management
• I .4 Thermal Management
• 1 I 5 Guidance, Navigation and Control
i.t.t Communications. Control and Health Management
1.17 LiFe Support
1.18 Environmental and Salerg Management
12 Vehicle Elernemrto Integration (Booster, Orbirer. ILl element. Planet or Moon DecenrlAsr
1.2 I Element ro element nrructwral arrachmenr
1.22 Element to element commnunicatron
1.23 Provide moorrorrog control 015.1. emrcitonrmrerrt between elements
1 24 Element ro Element Separation
3 Ground tnlrasrrocture Element(s)
1 3 t Flighr Element Processing
1 32 Pagload Elemenr Procennlnrg
I 33 Inrenrared Processing
34 Flight vrrd Ground TrulI'r T.c, 'i rr. vi iufmrg Managemenr
35 Ground Inlrasrructurc Supo:,'r a-'i V scagemenr
Ready NUM
Figure 2. Screen Shot of Third Level of Space Exploration System Functional Breakdown Structure.
'•81 - I I I I
= I t I
-J ' Ii I
'ont I
C
no
> =
.00 . . I
zzi5 . .
o
ZZZ
'ow o .Q
'
I
I
'I
I
I
I
I
I
I
sao
I I I
I i I
Lv. a, I I I
'oo '0 I I I
ur2_a,a, 'C 'C .5 I I
0 0 I
tn = u g , Gertenc Fu,ction De gcnp(ion (FBS 5thl6th I i I
I I
cID w u on CI on (0 an ..i Level)
VEHICLE
4kchrtectutal Concept Name . -
Vehicle Element )e.g.. Booster'. Orbiter, Payload element, repeat as needed for elements)
1.1.1 twfmme Structure & Mechanisms
1 .1.2 Propulsion
1 1,3 Power Management . . . . - _________ -
1 1,4 Therrrral Management
1 .1.5 Guidance. Navtgatrorl and Control _________________ -
1 1.6 Communications, Control and Health Management
INTERFACE 11.7 Lrte Support .
1 1.8 Environmental and Safely Management
Vehicle Elements lnteU'ation )Booster, Orbiter, TLI element, Planet or Moon Decent/Ascent element)
1.2.1 Element to element structural attaciment
1.2.2 Element to element commiwication
1.2.3 Provide morviorirrg & control of safe environment between elements -_____________
1,2.4 Element to Element Separatron _____________
GROU F t1 D
Ground Infrastructure Element)s( L I
1.3.1 Flight Element Processing , -
I_1,....._ I_I -
1 .3.2 Payload Element Processing I I 1
1
1.3.3 Integrated Processing ________ I -I I -
1.3.4 Flight and Ground TraffIc Control and Safety Management 1 11 1
1.3.5 Ground Infrastructure Support and Manament ' , ] 1' T1
As an example of the content in the spreadsheet, one third level function will be followed through its deeper
levels in the following figures. Figure 4 shows the fourth level functions decomposed from the third level function
"1.1.2 Propulsion". It was decomposed into five lower level functions.
1.1.2 Propulsion
1.1.2.1 Booster/Planetary Ascent Propulsion
The example continues and shows that one of the fourth level functions, "1.1.2.1 Booster/Planetary Ascent
Propulsion", was decomposed into 21 fifth level functions as shown in Figure 5.
Finally, Figure 6 shows the sixth level functions that resulted from the decomposition on the first six of the fifth
level functions shown in Figure 5.
9
American Institute of Aeronautics and Astronautics
45th AIAA/ASME./SAEIASEE Joint Propulsion Conference AIAA 2009-5344
2-5 August 2009. Denver. Colorado
CD E F G I HI
=
229 1.1.2 Propulsion
1.1.2.1 Booster/Planetary Ascent Propulsion ______ ______
1.1.2.1.1 Fill& Drain _____ ______
1.1.2.1.1.1 Oxidizer F&D system ______ ______
1.1.2.1.1.2 Fuel F&D system ______ ______
234 1.1.2.1.2 On-Board Propellant Storage ______ ______
1.1.2.1.2.1 Oxidizertank ______
______ 1.1.2.1.2.2 Fueltank ______
1.1.2.1.3 Cryogenic On-Board Propellant and hardware Conditioning for Engine Start ______ ______
238 ________ 1.1.2.1.3.1 Oxidizer bleed or bubbling system to provide thermal conditioning _______
________ 1.1.2.1.3.2 Fuel bleed and fluid circulating system to provide thermal conditioning _______
1.1.2.1.4 Storable Propellant Conditioning for Engine Start ______ _______
24i _______ 1.1.2.1.4.1 Engine oxidizer feed system fill & bleed ______ ______
242 ________ 1.1.2.1.4.2 Engine fuel feed system fill & bleed ______ _______
1.1.2.1.5 On-Board Purge ______ ______
244 _______ 1.1.2.1.5.1 Engine oxidizer system purge & conditioning to remove contamination ______
________ 1.1.2.1.5.2 Engine fuel system purge & conditioning to remove contamination _______
246 1.1.2.1.6 Pressurization ______ ______
1.1.2.1.6.1 Oxidizer tank pressurization ______________
1.1.2.1.6.2 Fueltank pressurization ______________
Looking at the sixth level functions in Figure 6, it would be possible to further decompose these functions to
even lower levels. In general, it will be possible to decompose functions to different levels for various fourth or fifth
level functions. How far this is done depends on the use of the FBS. For architectural LCC comparisons, the FBS
may only need to be taken to level three or four. On the other hand, for WBS development, or for technology gap
analysis, or for defining functional requirements, it may be desirable to take the functional decomposition as far as
possible.
VII. Conclusions
Low life cycle costs for space transportation systems have not been achieved in the past. Part of the cause for this
failure is that:
1. architecture options have not been compared on the same level,
a. derivative options often have most functions well defined,
b. new options often leave out needed functions,
c. but the options are compared anyway;
3. not all mission required functions have been identified at program initiation,
a. late identification added unexpected costs,
b. late identification prevents overall system optimization;
2. there has been a sub-optimal level of functional integration across elements,
a. in architectures to be compared,
b. in the chosen architecture,
i. leads to "stove-piping",
ii. leads to lack of developed components and designs being available to later programs.
Part of the problem is the lack of good approaches and good tools to identify and assess all of the needed
functions of a specific mission.
Recommendations
The SPST recommends that a Functional Breakdown Structure be generated at the beginning of architecture
studies and comparisons, that the FBS be used to ensure the level comparison of architecture options, that the FBS
be used to fully identify all needed functions, and that the FBS be used to identify and manage functional integration
across elements within the architecture. The FBS can also be used to help generate the initial set of \VBS's. The
space transportation system for the exploration mission can be addressed with the FBS already developed by the
SPST.
10
American Institute of Aeronautics and Astronautics
45th AIAA/ASME/SAEJASEE Joint Pmpulsion Conference ATAA 2009-5344
2-5 August 2009, Denver, Colorado
References
Zapata, E., Levack, D. J. H., Rhodes, R, Robinson, J. W., "Shuttle Shortfalls and Lessons Learned for the Sustainment of
Human Space Exploration", 45th AIAAIASME/SAE/ASEE Joint Propulsion Conference, AIAA 2009-5346, August 2009.
2 McCleskey, C. M., NASA TP-2005-2 11519, Space Shuttle Operations and Infrastructure, A Systems Analysis of Design Root
Causes and Effects, NASA Kennedy Space Center, Florida, April 2005.
A Guide for the Design of Highly Reusable Space Transportation - Final Report, Space Propulsion Synergy Team (SPST),
August29, 1997.
"Penn, J. P., Odom, P. R., Rhodes, R., Dankhoff, W., "SPST Collaborative Prioritization of Advanced RLV Technologies
Derived from a Bottom-Up Process", 37th AIAAJASME/SAE/ASEE Joint Propulsion Conference, AIAA 2001-3983, July 2001.
Rhodes, R., Dankhoff, W., Donahue, B., Christensen, D., "Current Space Shuttle System Shortfalls Assessment," SPST Space
Propulsion Synergy Team report to NASA Headquarters Director of Strategic Investments, Sept 2005.
6
Buede, Dennis. M. The Engineering Design of Systems: Models and Methods. New York: John Wiley and Sons. 1999.
Shepard, Alan and Slayton, Deke. Moon Shot. The Inside Story of America's Race to the Moon. Turner Publishing, Inc.
Atlanta, GA 1994. Pg 339.
8 DoD. Mil HDBK 881, DoD Handbook-- Work Breakdown Structure.January 1998. AMSC N/A. para 1.4.1.
11
American Institute of Aeronautics and Astronautics