Content uploaded by Koki Ho
Author content
All content in this area was uploaded by Koki Ho on May 22, 2020
Content may be subject to copyright.
Space Architecture Design for Commercial Suitability: A Case Study in In-Situ
Resource Utilization Systems
Tristan Sarton du Jonchaya, Hao Chena, Anna Wiegerb,
Zoe Szajnfarberbc, and Koki Hoa
a Daniel Guggenheim School of Aerospace Engineering, Georgia Institute of Technology,
270 Ferst Dr, Atlanta, GA 30313, USA, kokiho@gatech.edu
b Elliott School of International Affairs, the George Washington University, 1957 E St
NW, Washington DC 20052, USA
c School of Engineering & Applied Science, the George Washington University,800 22nd
Street NW, Washington, DC 20052, USA
Abstract
Space Agencies are increasingly interested in stimulating non-traditional players to
participate more broadly in the space enterprise. Historically, high barriers to entry
in the space market have included challenges of working with the government
customer and high technical and financial risks associated with the complexity of
space exploration. More recently, agencies have used inducements (e.g., new
contracting mechanisms, access to testing facilities) to mitigate these barriers. While
these efforts mainly focused on reducing barriers to participation in existing
exploration architectures, this paper explores the viability of an alternative strategy.
Instead of providing inducements, which essentially subsidize participation, we
propose a new strategy for space agencies to treat “commercial suitability” as another
“-ility” and make it an explicit criterion of the initial architecture selection. This can
be an effective option when multiple equivalent architectures (as evaluated against
traditional cost, schedule, and performance measures) differ on their “commercial
suitability.” As a proof-of-concept for this strategy, we develop a case study with lunar
in-situ resource utilization plant systems as a basis for comparing the architectures
with dedicated mass-wise optimal design (selected using traditional architecting
strategies) vs. standardized mass-produced modular ISRU (selected using
commercially-suitable strategies). The results show that architecture selection that
considers commercial suitability upfront can achieve increased commercial
participation without compromising cost performance compared with the baseline
architecture. This serves as an existence proof for the potential value of this new
strategy.
1. Introduction
For decades, it has been the policy of the National Aeronautics and Space
Administration (NASA) to involve the commercial sector in space exploration. Around
the start of the shuttle era (1981 – 2011), efforts to involve the commercial sector
became more emphasized and eventually took two approaches. The first approach
focused on encouraging the development of capabilities that are only possible due to
the space program, applying capabilities developed in the space program to
terrestrial use, and fostering markets for those capabilities. The second approach
focused on promoting and maintaining the industrial base for space transportation
by committing NASA to be an anchor tenant (core customer) of companies developing
unique capabilities.
To lower barriers associated with the federal acquisition process, which affect all
commercial interactions, NASA has also explored alternative contracting mechanisms
that reduce the burden of working with the government. For example, in 2005, NASA
developed the Commercial Orbital Transportation Services (COTS) program, which
evolved into the Commercial Resupply Services and the Commercial Crew Program.
Based on the success of these programs, NASA established the Commercial Lunar
Payload Services (CLPS) in 2019. Both programs rely on NASA remaining an anchor
tenant of the services.
To date, NASA’s efforts to induce more commercial participation in space system
development have focused on reducing barriers to entry. These are important and
have resulted in successes to some extent including enabling acquisition of core
capabilities like a lunar lander through commercial acquisition methods. However,
these efforts focus on reducing barriers to, and motivating participation in, existing
exploration architectures.
In this paper, we offer an alternative, complementary, approach to inducing
commercial participation that relies on increasing the inherent value of participation.
If a government agency (e.g., NASA) explicitly considers commercial capabilities and
interests during the systems architecting process, and it chooses architectures that
create opportunities for commercial involvement, then commercial participation
should increase, independent of explicit inducements. We call this strategy
architecting for “Commercial suitability,” as a nod to other “-ilities” (e.g., flexibility [1-
3], survivability [4-5]) which are an active area of systems architecture research more
broadly. Commercial suitability can be an important criterion for initial architecture
selection. It is particularly effective when there exist multiple, otherwise equivalent,
“best” architectures (as evaluated against traditional cost, schedule, and performance
measures) that differ on their “commercial suitability.” The key question would be:
among the “best” options, are there differences in commercial suitability that can be
exploited?
As a proof of concept demonstration that this approach can be valuable, this paper
performs a case study comparing architecting strategies for the In-Situ Resource
Utilization (ISRU) systems for space exploration. We compare a baseline scenario –
wherein the ISRU plant is sized optimally for the given mission, as is typically done –
to one that is formulated with a commercial suitability perspective – where priority
is given to fixing the unit size (i.e., standardization) and modularizing the system to
enable more commercial contributions through a simpler interface and by
stimulating mass production. As a result, we find that the impacts of the learning
curve effects enabled by standardization in the commercially-suitable architecture
can potentially outweigh the inefficiencies from the (mass-wise) suboptimal sizes of
the modules. In fact, depending on the assumptions about the learning curve effects,
the standardized modular architecture can even reduce overall mission costs from
the baseline one, besides providing other system-level benefits such as redundancy.
Thus, this case study demonstrates the existence of the architectures that
endogenously motivate commercial participation, without compromising on other
traditional cost metrics for the agency.
We believe the proposed new space mission strategy can introduce a paradigm
shift from identifying the role of commercial players in a given space architecture to
designing a commercially-suitable space architecture where commercial players can
be leveraged optimally and thus stimulate future space commercialization.
The remainder of the paper is organized as follows. In Section 2, we review NASA’s
space commercialization strategy. In Section 3, we discuss the proposed concept of
commercial suitability and its associated strategy. Section 4 introduces a proof-of-
concept of the proposed strategy in the context of lunar ISRU. Finally, we conclude
this paper in Section 5 with a discussion of the proposed framework and future work.
2. Review of NASA’s Space Commercialization Strategy
Before introducing our concept of space architecture design for commercial
suitability, we review the history of NASA’s space commercialization strategies.
2.1 Brief History of NASA’s Commercial Engagement
Around the start of the Space Shuttle era (1981 – 2011), efforts to involve the
commercial sector became more emphasized and eventually took two approaches: 1)
directly investing in capabilities intended solely for use in space and 2) maintaining
the health of the space industrial base. Both approaches have required substantial
subsidies on the part of NASA. The emphasis was not on the incorporation of
commercial capabilities into space exploration endeavors beyond the defense-type
contracting mechanisms already in place [6].
The first approach focused on encouraging the development of capabilities only
possible because of the space program, applying capabilities developed in the space
program for terrestrial use, and fostering markets for those capabilities. In 1979,
NASA issued the “NASA Guidelines Regarding Early Usage of Space for Industrial
Purposes” [6]. As indicated by the title, the emphasis of these guidelines was to
encourage companies to use NASA resources. Specifically, the guidelines outlined
three incentives for companies: “Providing flight time on the Space Shuttle; Providing
technical advice, consultation, data, equipment, and facilities; and Entering into joint
research and demonstration programs with NASA and the private sector partner
funding their own efforts” [6]. These guidelines formed the basis of the 1984 National
Policy on the Commercial Use of Space, introduced under the Reagan administration.
The emphasis of the 1984 National Policy was on technology transfer – encouraging
companies to find terrestrial applications for technologies originally developed for
the space program [6]. This commercial use of space policy emphasized a flow of
capabilities from those developed for space to companies that sold them on Earth;
therefore, it promoted the development of new capabilities only possible because of
access to the space environment.
The second approach focused on promoting and maintaining the industrial base
for space transportation by committing NASA to be an anchor tenant of companies
developing unique capabilities. Just before the shuttle era ended in 2011, the current
14-page National Space Policy was published with a full-page section on commercial
space guidelines. The first bullet of the guidelines emphasizes that the government
shall “purchase and use commercial space capabilities and services to the maximum
practical extent” [7]. The space policy goes on to direct the government to refrain
from competing with the private sector [7]. As stated in the pricing policy for the
International Space Station (ISS) commercialization, “NASA is restricted from
competing with the U.S. private sector; therefore, if, at any point, a U.S. Entity is
available to provide any of these resources, NASA shall, to the best of its ability,
migrate the provision of such services to the non-U.S. government provider” [8]. The
National Space Policy “sets forth the goal of energizing and enhancing a competitive
U.S. domestic space industrial base” [9]. The existence of a marketplace for the space
industrial base is simply assumed in the 2010 National Space Policy.
2.2 Incentivizing Commercial Participation by Lowering Barriers
Federal acquisition regulations are notoriously difficult to work with to the point
that companies create segregated portions of their business in order to work with the
government. A 1992 paper on Department of Defense acquisitions of commercially
available aircraft and airframes [10] reveals that acquisition methods that include the
following discourage companies from selling to the government even when they
produce and sell the capability in the commercial market:
• requirements peculiar to the government
• excruciatingly detailed product descriptions
• accounting requirements imposed by the government
• layers of bureaucracy and oversight
• lack of uniformity in contracts
• existence of onerous clauses in the contracts
If they do sell to the government, companies often will create a segregated portion of
their business to protect against the costs of doing business with the government
affecting the commercial side of their business [10].
The regulatory and policy requirements of the federal acquisition regulations
create a barrier to entry that increases the cost of doing business with the
government [10]. The government is then put in the position of compensating for that
barrier by paying more for an otherwise commercially available capability and/or by
providing incentives to companies to sell to the government.
The 1992 paper, as well as others throughout the years, have argued that the
government should use a commercial-style acquisition process for commercially
available capabilities [10]. Such an acquisition process would be free of the onerous
requirements and overhead typical of the federal acquisition regulations, thus
lowering the barrier to entry for companies seeking to sell to the government and
lowering the cost for the government to acquire commercially available capabilities
[10].
In an effort to lower barriers associated with the federal acquisition process and
reduce the financial burden on the government, NASA developed the Commercial
Orbital Transportation Services (COTS) program in 2005 [11]. The success of these
programs has prompted a similar approach for the lunar exploration plans. Named
the Commercial Lunar Payload Services (CLPS), it calls for the contractor to “provide
all activities necessary to safely integrate, accommodate, transport, and operate NASA
payloads using contractor-provided assets, including launch vehicles, lunar lander
spacecraft, lunar surface systems, Earth re-entry vehicles, and associated resources”
[9]. The request leaves the means quite open so that the contractor is free to choose
how to develop or purchase the sub-capabilities. Similar to COTS and Commercial
Crew at their starts, CLPS is fostering capabilities that have not yet been proven but
that likely can be in the near term. In 2015, some of the NASA employees who had
developed the COTS idea proposed what essentially became CLPS in a conference
paper [11]. Calling the concept Lunar COTS or LCOTS, they highlighted the similarity
it had to COTS and the benefits it could bring. As with COTS and as is likely to be true
of CLPS, the assumption was that NASA would be an anchor tenant of the technologies
developed [11].
3. New Concept: Architecting for Commercial Suitability
To date, NASA’s efforts to induce more commercial participation in space system
development have focused on reducing barriers to entry. This paper proposes an
alternative, complementary approach that relies on increasing the inherent value of
participation through architecture selection. We contend that if NASA explicitly
considers commercial capabilities and interests during the systems architecting
process and chooses architectures that create opportunities for commercial
involvement, participation will increase, independent of explicit inducements. We call
this strategy architecting for “Commercial suitability,” as a nod to other “-ilities” (e.g.,
flexibility, survivability) which are an active area of systems architecture research
more broadly.
It may sound counterintuitive that systems’ architects would choose to constrain
their own design space, but it is not as radical as it seems. On the high complexity end
of the spectrum, system designers are keenly aware of available launch envelopes and
design their systems within those constraints. On the low complexity end of the
spectrum, there is clear guidance on which COTS electronics are suitable for use in
space systems based on their qualification and quality control procedures [12]. The
reason for constraining designs can be different: for example, in the launch vehicle
case, it would be prohibitively expensive to customize launch services for each
mission, so the relatively few available volumes are taken as fixed; in the context of
the electronic parts, the space sector is a relatively small customer and does not have
the market power to request specialized components. Thus, either space
organizations can work with what is available commercially or can take on the
substantial cost of developing a unique space electronics market – the result is the
same: designers take the available options as given, and architect around them.
With a strategy of architecting for commercial suitability, we suggest that this
approach be applied to a larger set of design decisions. Examples of such a
commercialization strategy include picking a specific size for a module (e.g., power
system) that is common to multiple missions and or has the same interface
characteristics as existing terrestrial applications. This is a way to leverage the
broader terrestrial market and enable economies of scale in the space market.
Commercial suitability can be an important criterion for initial architecture
selection and a particularly effective option when multiple equivalent architectures
differ only on their “commercial suitability.” Thus, a commercially-suitable
architecture is only chosen if the cost of the constraint (i.e., the impact of choosing to
design around a fixed standardized size or interface on performance) is balanced by
the potential benefit of the constraint (i.e., either the social value of increased
commercial participation, or the future benefits of reduced costs and or risk
reduction). We believe that there are many potential instances where the future gain
is sufficient to justify the choice on the basis of traditional performance measures. As
an existence proof of this concept, this paper selects an example – ISRU systems – and
demonstrates that with moderate learning effects, the choice to fix (i.e., standardize)
ISRU sizing can both create a commercial market and deliver value to a government
customer.
4. Proof-of-Concept Case Study: Lunar ISRU
As a proof of concept for the proposed space commercialization strategy, we
consider the case of lunar ISRU systems for exploration. ISRU is chosen because it is
a representative example of highly specialized technology that is typically optimally
customized for each mission; our goal is to explore the value of standardization in this
relatively conservative context.
Lunar ISRU has been considered as a promising technology for lunar and Mars
exploration, and past studies have evaluated the value of ISRU [13-16]. Many such
studies assumed that we can optimally choose the ISRU size dedicated to the mission.
However, the strategy of designing dedicated, one-time-only, ISRU systems for
each mission may not be the preferred strategy from the commercial-suitability
perspective. Rather, commercially-suitable architectures would give priority to fixing
the unit size (i.e., standardization) to simplify commercial contributions. The
standardized mass-producible ISRU units (i.e., also referred to as modules in this
paper) can drive down the cost, stimulating the commercialization of the technologies
for future missions beyond the particular mission of interest. For example, the
proposed architecture enables future companies to purchase standard ISRU units and
use it for their own missions, just like how CubeSats have become a new standard in
the satellite development industry.
Despite the benefit of commercially-suitable standardization, the government
agency would only choose this strategy when its benefit outweighs its cost. In this
case study, we show that the commercially-suitable standardization can, in fact,
potentially achieve a similar or lower cost compared with the traditional baseline
architecture, leveraging the cost reduction due to mass production, which we model
as learning curve effects.
4.1. Overview of the Case Study
We consider designing a lunar ISRU system to meet an annual demand of oxygen
production: [kg per year]. This value is considered to be 50,000 kg/year in this
study, which falls under the range of the propellant demand in cislunar space for Mars
space exploration [17]. We compare the following two approaches to achieve this
demand:
(i) the dedicated optimally sized ISRU systems, which represent the traditional
design approach (i.e., baseline): note that this solution is mass-wise
optimal;
(ii) the standardized modular ISRU systems, which represent the
commercially-suitable approach (i.e., new approach).
The metric for comparison is the cost. The cost and ISRU performance models are
discussed in the following.
4.1.1. ISRU Cost Model
As discussed above, the ISRU cost model is based on a linear model. Since the
nonrecurring cost for the ISRU plant applies to both cases equally, we do not consider
that in our comparison. Nonlinear cost models can be considered for a more realistic
analysis, which is beyond the scope of this paper.
For Case (i) evaluated with optimized ISRU, the specific cost is defined as [$/kg].
For Case (ii) with modular ISRU systems, the learning curve effects are
incorporated to represent the cost reduction caused by mass production [18]. Here,
a standardized ISRU module mass is chosen and used in common for all missions to
generate the given demand. The module cost [$/unit ISRU] is constructed using the
theoretical first unit specific cost [$/kg] (i.e., the baseline specific cost), the mass
of the standardized ISRU module [kg], the learning rate , and the number of ISRU
modules in the following way:
(1)
The total cost of the batch of produced COTS ISRU systems is then obtained by
multiplying the production cost rate by the number of produced ISRU systems:
.
(2)
In the considered case study, the theoretical first unit specific cost $40,000/kg
is used, and the module mass size is varied. This value of is consistent with Ref.
[19], which gives the range of the first unit specific cost of launch vehicles to be
between $3,600/kg and $10,700/kg and that of manned spacecraft to be between
$13,000/kg and $90,000/kg.
Fig. 1 illustrates the specific cost of ISRU modules () vs. the number of ISRU
modules () at the theoretical first unit specific cost of $40,000/kg with different
production learning rate . An important observation from Fig. 1 is that savings in
production cost are the most significant for the first few produced ISRU and for a
lower production learning rate.
Fig. 1. Learning-curve-based cost model for ISRU modules at the theoretical first
unit specific cost of $40,000/kg
4.1.2. ISRU Performance Model
The ISRU oxygen production model is modeled using a non-linear model derived
in Ref. [20] from Ref. [21]. This model is created based on subsystem-level design
models for lunar ISRU plant systems, including not only the chemical reactor but also
the excavator, hopper/feed system, storage, and power subsystems. The assumed
chemical process is the molten regolith electrolysis method.
The model represents the annual mass-specific oxygen production [kg
O2/year/kg ISRU] as a function of the ISRU mass . This function is as follows:
If .
If
.
(3)
This function is represented in Fig. 2. Note the monotonically-increasing nature of
this curve, which induces an increase in oxygen production efficiency as the ISRU gets
larger in size.
0
0.005
0.01
0.015
0.02
0.025
0.03
0.035
0.04
0.045
0 5 10 15 20
ISRU Specific Cost [$M/kg ISRU]
Number of ISRU Modules
r=0.8 r=0.85 r=0.9
Fig. 2. Non-linear ISRU oxygen production model
4.1.3. Integrating Cost and Performance Models
Using the aforementioned cost and performance models, we can analyze Cases (i)
and (ii) in the following way:
For Case (i), we need to find the ISRU plant size to satisfy the demand optimally
in terms of the mass. Therefore, we need to solve the following optimization problem:
.
(4)
Note that is a monotonically increasing function; thus, the solution of this
optimization is also the solution of .
With this and , we can compute the cost using the aforementioned cost model.
For Case (ii), we want to analyze the cost given the ISRU plant size , thus we need
to compute the number of needed ISRU plants to satisfy the demand . This can be
found in the following equation:
.
(5)
where is a ceiling function. Given this value of , we can evaluate the cost using the
cost model given the ISRU plant size . Additionally, we can vary the value of
further to find the optimal ISRU plant size .
4.2. Case Study Results
This subsection discusses the case study results. In the case study, we vary the
standardized ISRU plant module size (i.e., mass) for Case (ii), i.e., the commercially-
suitable standardization strategy, and compare its cost with the baseline Case (i).
Fig. 3 shows the main results: the relationship between the ISRU cost and the ISRU
plant module mass for Case (ii) for three different learning rates (0.8, 0.85, 0.9). The
baseline cost for Case (i) is also shown for comparison. The optimal solution for Case
(i) is $246.6M with an ISRU plant of 6,164.08 [kg]. We have a few findings from these
results.
0
1
2
3
4
5
6
7
8
9
0 1000 2000 3000 4000 5000 6000 7000 8000
Annual Specific Oxygen Production
[kg O2/year/kg ISRU]
ISRU Plant Mass [kg]
First, we can find the optimal ISRU plant module size given a learning rate for
Case (ii). For example, for , this optimal ISRU plant module size is around 1,000
kg. This can be explained as follows. On the one hand, when the ISRU plant module is too
large, we would have unnecessary extra ISRU capability, and thus the cost will be larger.
On the other hand, when the ISRU plant module is too small, each ISRU module can be
inefficient as can be seen in the monotonically-increasing ISRU performance model with
respect to the module mass (i.e., larger ISRU modules experience higher performance). An
additional important factor that impacts this tradeoff is the learning curve effects, which
prefers to have more small ISRU modules than fewer large modules. Given these above
factors, we would expect a convex curve to have an optimal ISRU plant mass size The
reason why the curves in Fig. 3 are not precisely convex is because of the integer number
of ISRU modules. For example, there are jumps in the cost curve when the ISRU plant
module mass is around 3,000 [kg] and around 6,000 [kg], which corresponds to when the
number of needed modules dropped from 3 to 2 and from 2 to 1, respectively. Also, note
that, beyond this pure analytically optimal ISRU plant size, the actual standard size can
also be determined with consideration of existing similar technologies (e.g., mining
technologies) so that these existing commercial players can naturally be incentivized to
participate in the mission; this analysis is left for future work.
1
A more important finding is that the proposed standardization can achieve a similar
or even lower cost than the mass-wise optimal solution. For example, Case (ii) achieves
a lower cost than Case (i) when the ISRU plant size is 2,000 [kg] at the learning rate of
0.85 or 0.8. This is because the impacts of the learning curve effects outweigh the
inefficiencies due to the mass-wise suboptimal modular size, resulting in a reduced
cost. In this case, the system has four ISRU plant modules, each being 2,000 [kg], rather
than one large plant; the learning curve effects due to producing four modules outweigh
the extra cost of modularization, besides providing other system-level benefits such as
redundancy. This cost reduction is expected to promote further mass production and thus
stimulate commercial participation in this enterprise. For example, the companies can
leverage this reduced cost of the standardized ISRU units for other government or
privately-funded lunar, Mars, and even asteroid missions beyond this particular mission.
This case study demonstrates the feasibility of our strategy to achieve
commercially-suitable architectures with little or no sacrifice in cost. By choosing
such architectures with standardization and thus explicitly creating opportunities for
commercial involvement, the space agency can attract commercial players to
participate in the enterprise. Note that we do not claim that the obtained ISRU designs
from this simple case study are the optimal architecture to pursue (e.g., modularizing
only part of the subsystems may be more effective); rather, our goal is to provide a
proof-of-concept to show the promising nature of the concept of commercial
suitability.
1
The 1,000-2,000kg ISRU module size is consistent with NASA’s plan on ISRU [23], and thus is a
realistic module sizing.
Fig. 3. ISRU Cost vs. ISRU Plant Module Mass for Case (ii). The Case (i) baseline cost
is added for reference.
5. Discussion and Future Work
This paper proposes a new way to think about increasing (and broadening)
commercial involvement in the space industry. Instead of encouraging participation
through monetary subsidies and/or favorable contracting mechanisms, it suggests
that commercial entities can be endogenously induced to participate by creating
value for them through architecture selection choices. Architecting for commercial
suitability is offered as a strategy to achieve this.
We illustrate the potential feasibility of this approach with a case study of ISRU
systems by demonstrating an example of a commercially-suitable architecture (i.e.,
standardized ISRU) that can also potentially achieve a similar cost or even reduce the
cost compared with the baseline architecture. Thus, we prove that there exist such
architectures that induce commercial participation without negatively impacting
traditional measures of effectiveness. More broadly, this style of analysis can be used
to identify opportunities to apply selective constraints during the architecting
process that can stimulate broader commercial participation in future missions.
Now that the viability of this approach has been demonstrated, we can address
specific strategies for implementing it. Although those specifics are left for future
work, our general vision is to work within existing architecting frameworks in two
main ways: First, generic strategies for modularity, standardization, and
commonality can be applied to generate more potentially commercially-suitable
options in the tradespace. Second, scorecards can be developed to aid in the screening
for commercial suitability during the analysis of alternatives. In this initial proof-of-
concept, particular architectures are coded as commercially suitable or not, and then
compared on standard measures (e.g., cost). A more sophisticated analysis can
require a more complete assessment of the value of broader commercial
0
50
100
150
200
250
300
350
400
450
500
0 1000 2000 3000 4000 5000 6000 7000 8000
ISRU Cost [$M]
ISRU Plant Module Mass [kg]
Case (i) Baseline Case (ii) r=0.8
Case (ii) r=0.85 Case (ii) r=0.9
participation. For example, decision-makers would want to weigh a small increase in
cost against potentially stimulating secondary markets. Ref. [22] presents a potential
scorecard-based approach.
Space Agencies are increasingly interested in stimulating non-traditional players
to participate more broadly in the space enterprise, and a novel alternative strategy
for doing so is proposed in this work. Our hope is that this framework will enable
future research to explore strategies for generating commercially-suitable
architectures and also valuing their advantages during the architecting process.
Acknowledgment
This material is partially based upon work supported by the funding from NASA
(80NSSC17K0329) awarded to the University of Illinois and George Washington
University, where this work was initiated. Any opinions, findings, and conclusions or
recommendations expressed in this material are those of the authors and do not
necessarily reflect the views of NASA. The authors appreciate Onalli Gunasekara for
proofreading this paper.
References
[1] A. M. Ross, D. H. Rhodes, and D. E. Hastings. Defining changeability: Reconciling
flexibility, adaptability, scalability, modifiability, and robustness for maintaining
system lifecycle value, Systems Engineering 11.3 (2008): 246-262.
[2] E. T. Ryan, D. R. Jacques, and J. M. Colombi. An ontological framework for clarifying
flexibility‐related terminology via literature survey, Systems Engineering 16.1
(2013): 99-110.
[3] M-A. Cardin, Enabling flexibility in engineering systems: a taxonomy of
procedures and a design framework, Journal of Mechanical Design 136.1 (2014).
[4] M.G. Richards, D. E. Hastings, D. H. Rhodes, A. M. Ross, and A. L. Weigel, Design for
survivability: concept generation and evaluation in dynamic tradespace exploration,
2nd International Symposium on Engineering Systems. 2009.
[5] M. G. Richards, A. M. Ross, D. H. Rhodes, and D. E. Hastings, Metrics for evaluating
survivability in dynamic multi-attribute tradespace exploration. Journal of Spacecraft
and Rockets, 46(5) (2009): 1049-1064.
[6] J. Rumerman (Compiler), Chapter 5: Commercial Programs, in: NASA Historical
Data Books (SP-4012) Volume VI, NASA History Office, Office of Policy and Plans,
Washington, D.C., 2000, pp. 355–362.
[7] Office of the President of the United States, National Space Policy of the United
States of America, 28 June 2010.
[8] D. Elburn, Commercial and Marketing Pricing Policy, 7 June 2019,
https://www.nasa.gov/leo-economy/commercial-use/pricing-policy, (accessed 3
October 2019).
[9] National Aeronautics and Space Administration (NASA) Headquarters (HQ),
Commercial Lunar Payload Services (CLPS) Justification for Other than Full and Open
Competition (JOFOC), 13 December 2018.
[10] D. W. Humerick, S. H. Minnich, Applying Commercial Style Acquisition Practices
to the Procurement of Commercially Available Aircraft, M.S. Thesis, Air Force Institute
of Technology 1992.
[11] A. F. Zuniga, D. Rasky, R. B. Pittman, Lunar COTS: An Economical and Sustainable
Approach to Reaching Mars, AIAA SPACE 2015 Conference & Exposition, Pasadena,
California, 2015.
[12] M. White, Commercial Off-the-Shelf (COTS) Parts Risk & Reliability User &
Application Guide, Jet Propulsion Laboratory, California Institute of Technology,
Pasadena, CA, 2017.
[13] T. Ishimatsu, O. L. de Weck, J. A. Hoffman, Y. Ohkami, and R. Shishko, Generalized
Multicommodity Network Flow Model for the Earth-Moon–Mars Logistics System,
Journal of Spacecraft and Rockets, 53(1) (2016) 25-38.
[14] K. Ho, Dynamic network modeling for spaceflight logistics with time-expanded
networks, Ph.D. Thesis, Aeronautics and Astronautics Dept., MIT, Cambridge, MA,
2015.
[15] H. Chen, H. Lee, and K. Ho, Space Transportation System and Mission Planning
for Regular Interplanetary Missions, Journal of Spacecraft and Rockets, 56(1) (2019)
12-20.
[16] H. Chen, T. Sarton du Jonchay, L. Hou and K. Ho, Multi-Fidelity Space Mission
Planning and Space Infrastructure Design framework for Space Resource Logistics,
AIAA Propulsion & Energy Forum 2019, Indianapolis, IN, no. AIAA 2019-4134, Sep.
2019.
[17] C. Jones, R. Merrill, and E. McVay, Cis-Lunar Reusable In-Space Transportation
Architecture for the Evolvable Mars Campaign, AIAA SPACE 2016 Conference &
Exposition, Long Beach, California, 2016.
[18] J. Wertz, D. Everett, J. Puschell, “Space Mission Engineering: The New SMAD,”
Microcosm Press, 2011.
[19] Preliminary cost analysis, Principles of space systems design,
https://spacecraft.ssl.umd.edu/old_site/academics/483F02/09_costing_2002.pdf,
(accessed 16.06.2019).
[20] H. Chen and K. Ho, Integrated Space Logistics Mission Planning and Spacecraft
Design with Mixed-Integer Nonlinear Programming, Journal of Spacecraft and
Rockets, 55(2) (2018) 365-381.
[21] S. S. Schreiner, Molten Regolith Electrolysis Reactor Modeling and Optimization
of In-Situ Resource Utilization Systems, M.S. Thesis, Aeronautics and Astronautics
Dept., MIT, Cambridge, MA, 2015.
[22] A. Wieger, H. Chen, T. Sarton du Jonchay, K. Ho, and Z. Szajnfarber, An Approach
to Endogenously Incentivizing Commercial Participation through System
Architecture Choices, International Astronautical Congress, Washington DC, Oct.
2019.
[23] D. Linne, J. Kleinhenz, L. Sibille, J. Schuler, N. Suzuki, L. Moore, S. Oleson, A.
Colozza, E. Turnbull, “Oxygen Production System for Refueling Human Landing
System Elements,” 10th Joint Meeting of the Space Resources Roundtable / Planetary
and Terrestrial Mining and Sciences Symposium, 2019.
