On January 14, 2004, President George W. Bush presented the nation with a bold new initiative to “explore space and extend a human presence across our solar system…using existing programs and personnel…one mission, one voyage, one landing at a time.” (Bush, 2004) NASA was charged with the task of developing a sustainable and affordable human space exploration program with the initial objective of returning a human presence to the Moon by the year 2020. The directive thus raises two broad engineering questions: First, what is the purpose of an exploration system, and how one evaluates its performance. Second, how does one architect a sustainable space exploration system? The following report makes the case that the primary purpose of an exploration system is the delivery of knowledge to the stakeholders, and that the design should be evaluated with respect to knowledge.

Return to Table of Contents

1. Introduction

On January 14, 2004 President George W. Bush presented the nation with a new vision for space. The National Aeronautics and Space Administration (NASA) will develop a sustainable human space exploration program taking humans back to the Moon by 2020, and eventually to Mars and beyond (Bush, 2004). The vision, and plan that goes with it, calls for the completion of the ISS, the retirement of the Space Shuttle by 2010, and the development of a new Crew Exploration Vehicle (CEV). Bush’s vision provides a bold push towards mankind’s traversing of the solar system. The following report, representing the culmination of MIT’s 2004 spring 16.89 graduate design class, presents a design methodology and conceptual tools to facilitate the achievement of this vision. It addresses two critical questions facing the space community: What is sustainability in the context of space systems? How can sustainability be provided for during conceptual design? The following report addresses these questions. In doing so, it demonstrates that an exploration program is by definition a knowledge acquisition and transfer system, and it presents a process by which one may design for sustainability.

The goal of exploration is knowledge

While the motivation behind exploration has varied throughout history, the primary function of any “exploration system” has been to discover the unknown, to gain knowledge. Some of the more common ways to gain knowledge have been through the use of visual, electrical, or physical transportation of information. A simple example of a space knowledge transfer system is the human eye. The human eye gathers knowledge in the form of light. Several hundred years ago mankind developed the telescope in a hope to improve upon the amount of knowledge delivered to the eye through the discovery of magnification. The magnification of objects resulted in a higher order of knowledge resolution and consequently more information about space was discovered.

More recently mankind has sent satellites and drones into the solar system, with sensors that can gather information unattainable by the human eye alone. Information gathered by these systems is sent back to Earth through the use of electrical transmissions where it is turn into knowledge. A number of characteristics increases the “knowledge resolution” of these satellites and drones compared to telescopes, including: Shorter distance between optics and target, physical contact, sample return, in-situ analysis, etc. It is noteworthy that order to achieve this higher of knowledge resolution, mankind had move beyond light as the sole transfer-mechanism, to in-situ measurement and mass transport. Future exploration systems must necessarily follow this trend, exploiting the duality between mass and knowledge transfer, with one critical improvement--humans will provide degree of knowledge resolution previously unimaginable with satellites, drones, and telescopes alone.

No matter the form of the space exploration system (human eye, telescope, robotic probe, or human contact), the end product of the exploration system is knowledge. Currently, the majority of the work being completed on NASA’s new initiative is directed towards a new exploration vehicle. The class believes that any new space vehicle developed by NASA must be designed with an understanding that it will be but one tool in system whose ultimate function is to gather and transfer knowledge in space and on Earth.

To say that an exploration system must deliver knowledge to achieve its goal is to recognize that while mass transport enables exploration, the ultimate success of an expedition depends on the acquisition, communication, and synthesis of visual imagery, scientific data, and human experience to key stakeholders. This suggests revaluing traditional space system characteristics and trades to account for the demands of knowledge acquisition and delivery. Further, in order to make clear decisions about system capabilities and mission goals, attributes of knowledge must be categorized and valued in accordance with stakeholder needs. System designers must have a firm grasp of the knowledge delivery process, and establish how it will occur at each point in the system’s lifecycle.

Return to Table of Contents

Sustainability in the Design Process

Before knowledge can be incorporated into system valuation and trades, however, there must be a clear understanding of what is a sustainable space system and how can this can be addressed during conceptual design? Current space system design methods are not geared towards enhancing “sustainability.” Traditionally, they have focused on developing requirements, conducting trades based on assumptions about the future, and then optimizing the system with regard to some metric. Results are commonly single point designs optimized for single missions.

While such methods have proven adequate for low-frequency missions, they rely on assumptions about an uncertain future. A design that is optimal at one point in time may become less optimal in the future. Due to the expected duration of the new exploration initiative, major investments should note be made based on unverified assumptions. The new exploration system should be designed so that it can respond to changes in the future. The approach to design described in this report addresses this problem. Using an iterative process, and emerging system valuation tools, it creates a rigorous development strategy which is flexible and robust to environmental changes.

Chapter two proposes a definition of sustainability. Drawing from recent scholarship and historical examples, it argues that sustainable exploration programs must first and foremost have the capability to manage various kinds of uncertainty, including policy, budgetary, technical, and logistical changes. Conceptual designs must provide system operators with the ability to anticipate and capitalize on emerging opportunities and positive feedback loops while simultaneously adapting to changing value-structures and external circumstances.

Properties that enable sustainability have been termed flexibility, extensibility, robustness, and commonality. Much recent scholarship has addressed the need to rigorously value these system properties for the purposes of design. Generally, these properties translate to formal architectural attributes, such as modularity and platforming, as well as operational attributes such as staged deployment and spiral development. Chapter three defines these terms in the context of space systems, and presents methods for their formalization in system architecture.

There are two ways in which flexibility and extensibility are introduced and evaluated: mathematical evaluation methods and architecture design considerations. The mathematical evaluation methods used are based upon decision analysis, real options theory, and scenario planning. The architectural design considerations are commonality, scalable systems, and modularity. Both methods evaluate a given system based on the resulting value of knowledge delivered by the system. Notice that the system is not evaluated on cost or mass, but on knowledge, which is the primary purpose of an exploration system.

A major aspect of this study involves identifying a process to combine these properties and methods can be systematically incorporated into system design. Part of the solution involves creating a strategy, rather than a point design, that can react to change. Chapter 4 presents an example strategy, or “baseline,” which was conceived through an iterative process of design, needs mapping, and synthesis of sub-strategies. Sub-strategies consist of small, medium, and large Moon and Mars expeditions, each designed with principles of extensibility such as commonality and staged deployment. Individually, these missions are rough “point-designs.” However, major architectural decisions in each reflect anticipation of gradually increasing mission scale, and eventual transit to Mars.

Return to Table of Contents

After completing the sub-strategies, areas of functional commonality and uniqueness can be anticipated across the system, and architectural forms refined appropriately. The resulting forms and operations can then be synthesized into an integrated life-cycle strategy, with options for reacting to uncertainty. The following schematic illustrates the design process used:

In developing the integrated baseline, commonality trades at the formal and operational level become necessary. Chapter 5 details such trade studies and their results.

Once the final version of the baseline strategy and associated trades has been developed, more rigorous tools may be applied to determine when, and under which circumstances different design options become valuable. For example, the decision to transit through the Earth-Moon Lagrangian Libration Point 1 (EM-L1) while en route to the Moon may not be optimal for a single mission to the lunar equatorial region. However, if the frequency of non-equatorial lunar missions is sufficiently high, the option of utilizing EM-L1 becomes increasingly valuable. Tools, including modified forms of real options valuation, can inform trade studies of this sort, resulting in up-front design decisions that drastically reduce life-cycle cost and increase system flexibility.

Chapter 6 introduces such tools and methods. Scenario planning is applied to the integrated strategy to examine how the system can react to environmental changes. Adjustments are then suggested, based upon the baseline’s reaction to the scenarios. Decision analysis and Real Options analysis techniques are also used to determine at what point time-critical decisions should be made in the execution of the baseline strategy, and which investments should be made now to allow for the option of adapting to future uncertainty.

Return to Table of Contents

2. Intro to Sustainability

What exactly is a sustainable exploration program? In one sense, the answer is rather straightforward. To “sustain” means literally: to maintain in existence, to provide for, to support from below (Dictionary.com website). At the programmatic level, an exploration system will be maintained in existence so long as it is funded, and it will be funded provided it meets the needs of key stakeholders, members of Congress, the Administration, and ultimately the American people. Realistically, however, system designers must recognize that these needs themselves will change. A multi-year, multi-billion dollar program in the US Government must expect to face a great deal of uncertainty with respect to objectives, budget allocations, and technical performance.

In order for an exploration system to be sustainable, then, it must be able to operate in an environment of considerable uncertainty throughout its life-cycle. Designing for sustainability implies identifying sources of uncertainty and managing them through up-front system attributes. Various terms have been used to describe such system attributes, including: flexibility, robustness, and extensibility.

While a large complex system must react to changing environments in order to be sustainable, technological aspects of systems can themselves impact the environment. Once in development and operation, a multi-billion-dollar system will mediate political interests, organizational decisions, and technical alternatives, creating potential sources of stability and positive feedback-loops, as well as sources of uncertainty. Early decisions that create high switching costs or large infrastructure sites, can “lock-in” architectural configurations and influence the objectives and development path of later systems (Klein, 2000). A sustainable design will be one in which, to the greatest extent possible, the dynamics behind political, technical, and financial sources of stability support, rather than hinder, system development and operations.

The following chapter identifies three kinds of sustainability, and relates these to formal system attributes. It reviews current thinking about flexibility and extensibility, and their relation to architectural form. The chapter concludes with a historical investigation of Antarctic exploration, drawing lessons for the sustainability of exploration programs.

Return to Table of Contents

2.1 Elements of Sustainability

It is increasingly evident that large, complex, technological systems cannot be conceived independently from the political, economic, and organizational environment in which they operate. While at a technical level, exploration is dependant on continuous and reliable logistical support, at a programmatic level, political and organizational factors greatly affect sustainability. With space activities in particular, motivations and objectives can change rapidly compared to system life-cycles, increasing the impact of political and organizational issues on system development and use. A sustainable space exploration system will successfully mediate and react to political, organizational, and technical uncertainty, and also exploit, to the extent possible, sources of “stability” that arise from the interaction of these factors.

2.1.1 Policy Sustainability

Policy uncertainty can take the form of changes in objectives or the regulatory environment in which a system must operate. It stems from the dynamic nature of the US government, and the need for space systems to suit both national/strategic and political/tactical interests. Government programs are re-assessed on a yearly basis in terms of national priorities and, in some cases, performance. Changes in the political and geopolitical environment can alter the perception of the value of exploration activities. An important aspect of policy sustainability is thus the ability to maintain relevance, and continue operations, in the face of shifting objectives and regulatory environments.

To take one example, while the decision to build the Space Station Freedom was motivated largely by Cold War concerns, the fall of the Berlin Wall transformed the ailing project into a symbol for international peace and cooperation (Wikipedia, 2004). To the extent possible, system designers should consider the implications of such changes for system operation. If a policy decision to focus on Mars rather than the Moon is likely in the near term, current designs should be extensible to both objectives. Similarly, if international cooperation is based on uncertain agreements, alternatives to international participation on the critical path of development should be available.

2.1.2 Budgetary Sustainability

Shifting political priorities also create changes in funding. During its years of development and operations, a programs budget may oscillate unpredictably. Figure 1 illustrates how NASA’s budget fluctuates over time.

Figure 1: NASA budgetary fluctuations in 1996 dollars (courtesy http://history.nasa.gov)

A flexible system will maintain exploration capability even in the face of budgetary fluctuations, whether through changes in schedule, scale of operations, or by other means.

2.1.3 Organizational Sustainability

Recent scholarship has investigated the relationship between organizational structure and technical design. Charles Perrow (1984) has characterized socio-technical systems in terms of their dynamics and complexity, drawing conclusions for system safety and reliability. He defines space systems as highly coupled, nonlinear, and complex. Organizational structure and technical complexity can impact system reliability by creating “quite erroneous worlds in [the] minds” of system operators and managers. (Perrow, 1984)

Diane (1996) Vaughan furthers this understanding, suggesting that “the microscopic world of daily decisions” can create almost imperceptible changes in organizational culture over time, with important consequences for safety. Her term, the “normalization of deviance,” encompasses the way in which expectations can change and aberrations become accepted, through continual exposure to anomalies. Organizational structure, which impacts daily decisions, plays an important role in system performance and reliability, and thus sustainability.

A space system will be sustainable from an organizational perspective, then, if the technological system and management structure are designed together to minimize organizational drift and normalization of deviance.

2.1.4 Technical Sustainability

Technical sustainability refers to system performance, reliability, and the potential infusion of new technologies. An exploration system must support and maintain human and robotic activity at various fronts of exploration, and incorporate technological advances to continuously improve system performance without major operational changes. Further, any highly complex system is likely to fail at some point during its life cycle. A sustainable system will be one that is robust to failures, both small and large.

An important factor related to technical sustainability is risk tolerance. Risk tolerance can be divided into three main areas:

Development risk: during design, test integration of architecture components
Planning risk: willingness to exploit more or less of known system margins while planning an exploration mission
Operations risk: willingness to take risk during operations.

By definition, risk-free exploration does not exist. System designers must balance the risk associated with architectural form, schedule, and operations, in order to achieve system objectives. Risk tolerance can change throughout a system life cycle, and thus change how a given system is operated.

Return to Table of Contents

2.2 Sustainable Exploration Systems – Dynamics

While each of the three domains above impacts the development and operations of complex systems in different ways, they are closely interrelated. The dynamic relationship between the three has important ramifications for sustainability. The relationship between these three broad domains is shown in Figure 2.

Figure 2: Interaction of political, organizational, and technical factors

The Columbia Shuttle Accident Report (CAIB) repeatedly stresses the adverse affects that broader issues such as indecisive national leadership, increasingly stretched budgets, and continued mischaracterization of Shuttle capabilities have had on NASA’s organizational and safety culture.

Conversely, Hans Klein has suggested that the characteristics of a technological system and development program can facilitate or impede coalition politics, thereby reducing or exacerbating conflicts between politics and program administration (Klein, 2000). Technology and politics are linked when program administrators translate political forces into design requirements. Further, once developed, a given system architecture together with its supporting facilities can become “locked-in” and perpetuated through later designs. The space shuttle, for example, made use of facilities designed partly as the result of short-term political wrangling conducted during the Apollo era (p. 319 [ESS1] ).

Annalisa Weigel and Daniel Hastings have similarly investigated the interrelation between technical design and political change (2003). Weigel and Hastings stress that space transportation infrastructures are affected as much by political considerations as technical problems. It is thus imperative to understand the coupling of both domains if a system is to operate successfully in the “politico-technical” arena. Weigel presents a framework to understand how policy directives couple with technical parameters. Figure 3 is an “influence diagram” used to illustrate such coupling.

Figure 3: Translating policy parameter affects into the technical domain: an influence diagram (courtesy, Weigel and Hastings, 2003)

At a different level, as a later section of this chapter notes, the interplay between news, politics, and technical development was an important factor in the evolution of Antarctic exploration. In this respect, designing for sustainability implies understanding how various design decisions can lead to organizational and political dynamics that may improve or impeded the flexibility of the system.

Return to Table of Contents

2.3 Sustainability, Flexibility, Robustness

A sustainable system will have attributes that allow it to cope with, or mediate, various forms of uncertainty throughout its life-cycle. Many terms have been used to define characteristics which give systems these properties. They include flexibility, robustness, and extensibility. But what are the relationships between these terms?

In many ways this is simply a question of definition. Flexibility can be defined as the ability of a system to change or be used differently than intended after it is initially fielded. Flexibility can be intentional, but is often unintentional such as in the case of the B-52 or the use of the LM as a “life boat”. The speed with which a system reacts to change is a measure of agility. Extensibility is a particular kind of flexibility. Conversely, robustness is the property of a system that allows it to be insensitive to change. A system is robust if it continues to deliver value in changing circumstances.

All of these “ilities” are enabled by attributes of architectural form. The follow schematic illustrates how the various concepts relate to each other:

Example Architectural Form

Return to Table of Contents

2.4 Extensibility – An Enabler of Sustainability

Extensibility is defined as “the property that new elements can be added to a system in such a way as to alter the value delivered.” (Crawley, 2003) Designing systems to be extensible drives life cycle cost down through anticipating future goal and environmental changes and then translating this understanding into upfront system design actions aimed at minimizing overall life-cycle cost. Extensibility addresses both known and unknown future changes, with expected payback being variable, based on the certainty and magnitude of the anticipated change, along with the cost associated with making the system extensible.

Designing systems for extensibility requires a fundamental shift in the way design decisions are made, a shift from near optimal fulfillment of immediate requirements at minimal cost, to minimizing life cycle cost, maximizing life-cycle performance, etc. In other words, an extensible design will not be the highest performing design when compared to a point design optimized for a given set of capabilities- a penalty is placed on ultimate system performance in order to increase life-cycle value. An extensible design will not be the lowest cost design under the same conditions either. The advantages of an extensible design are only realized in the context of multiple generations of the system. New metrics must be implemented for valuing the benefits of extensibility. In addition, a culture shift must occur from near term to longer-term expectations of success.

The large investment associated with complex systems dictates the need for an evolutionary growth path, although not all elements of the system undergo the same degree of change. Therefore, it is important to invest “extensibility dollars” only where needed. Investing in extensibility provides an option for future change. As an example, an in-space crewed exploration vehicle could be designed for extensibility in terms of number of crew supported and days of support through decoupling of living quarters with the command and control portion of the spacecraft. While the initial need may be support of a four-person crew for two weeks, this need may extend to support of six people for nine months. Clearly, using the same vehicle for both missions would unduly penalize the shorter mission while design of two separate vehicles would result in high costs associated with development of redundant functions such as the command and control functions. Separating the habitat functions from command functions through creation of two modules and a common interface, for instance, would enable the habitation portion of the spacecraft to be easily modified. If the change is executed, the implementation of the change is expected to cost less than if the option had not been put into place. If not executed, the extensibility feature represents wasted resources in terms of the expense to implement, reproduce and support the unused feature.

Several concepts overlap almost directly with extensibility- staged deployment, and spiral/incremental development. Staged deployment seeks to match demand and supply through scaled rollout of a system. Expenses are delayed until a later date, reducing the net present value of the expense and increasing the certainty of the need, at the time of the expense. De Weck et al. (2004) describe staged deployment as a potential alternative to full deployment of the Iridium communications satellite network, with the potential benefit being lower investment in order to start operations. Additional satellites could have been deployed as demand increased. While Iridium was ultimately displaced from most of the expected market due to widespread cellular coverage, the lost investment could have been significantly reduced.

Like staged deployment, spiral development (Figure 4) is also an incremental method of deploying new systems and their capabilities in a flexible manner. Initial capabilities are selected based on prioritized goals, enabling quick deployment of high priority capabilities. Additional iterations of the process focus on deploying lower priority capabilities and addressing newly discovered needs. The result is quick deployment of primary capabilities combined with risk reduction through decision delay that enables incorporation of current technology into new stages and shifts in strategy as needs become clearer (time advances).

Figure 4: Boehm's model of spiral development (picture from Boehm, 1988)

Return to Table of Contents

2.4.1 Reasons for Extensibility

Extensibility reduces overall life-cycle cost and/or increases life-cycle performance through a number of difference paths. Several are listed below, along with brief descriptions.

2.4.1.1 Management of Technology Obsolescence

As the lifetime of a system grows, the rate of change of technology is increasingly mismatched with the scale of system replacements. Within a system, different modules have different rates of technological change. Charles Fine (1998) uses the term “clockspeed” to describe the rate of change and to highlight the differences between rates of change. Extensible systems allow for management of technological change within the system. As an example, consider a vehicle, such as a spacecraft. While structural technology may undergo significant improvements on the timescale of a decade or more; control system components, especially the electronic elements such as logic chips, undergo significant change on an annual basis. Designing a system to accommodate varying clockspeeds enables the design to evolve over time. One method for accommodating technological change is through grouping components with similar rates of change into modules, therefore, enabling easy replacement of the module, with minimal impact to other areas. Ease of change leads to the ability to keep a system modernized.

2.4.1.2 Risk Mitigation

Delaying decisions improves the likelihood of making a correct decision. While delay can cripple a program if not handled properly, the result of effective use of delay is confidence in decision-making.

2.4.1.3 Policy Fluctuation Robustness

Extensibility is beneficial in the face of the uncertainties produced by the policy domain, and the resulting budget fluctuations. The potential for a change in President occurs once every four years, a timescale much shorter than that of an exploration program. Given the mismatch in timescales, it is critical that achievement of intermediate milestones provides lasting value, a foundation for future work.

2.4.2 Describing Extensibility

Methods are needed for describing what an extensible system is and how the extensibility is achieved. Ultimately, the metrics and descriptions must be quantifiable to enable trades to be made between designs and design options. Which system is more extensible? How extensible is the system?

One view of the evolution of a system over time is a consideration of the relationship between available capabilities and required capabilities; in other words, a type of supply and demand curve. Figure 5 provides a notional view of this concept. The system needs over time are represented as a continuous curve. While the system needs curve may in fact be discrete, the aim here is to highlight the high degree of changing need in relation to the ability of the system to change. The design points represent the available capability levels. From a system performance standpoint, the ideal available capability would be a direct overlay over the needs curve. While the ideal curve cannot be reached due to practical considerations such as the cost of each change (engineering, deployment, etc.), the ideal curve can be approached through the creation of an extensible architecture. This view is closely related to previous work in the area of staged deployment. (de Weck et al., 2004).

Figure 5: Change in system need and capability over time

The relationship between the supply and demand “curves” is an important one. As Figure 5 illustrates, a system that is overly capable is inefficient. More dollars and time have been spent on unneeded functionality, at the given point in time. The reverse situation means that the system is not meeting needs, also a problem. As an example, consider the transition from Design 2 (D2) to Design 3 (D3). Before the transition, needs aren’t met by capabilities, while after the transition, the system is over-designed, as would be expected immediately after an improvement. Also note the transition from D3 to D4. While this transition was not required to meet new capabilities, since needs have actually decreased, the change was made in order to maintain design efficiency.

In order to analyze the evolution of a system over time, a well-defined method of describing change is needed. This void can be filled by a series of operators, such as those defined by Baldwin and Clark (2000):

Splitting (into two or more modules)
Substituting- replace module with a different one
Augmenting (adding a module)
Excluding- removal of a module from the system
Inverting- creation of new design rules
Porting- use module in another system

The above operators can be used to perform all module-level operations. As was mentioned in the previous section, it is critical to realize that evolution is synonymous with adaptation or change, not addition. Continuous adaptation to changing conditions may mean eliminating functionality that is no longer needed; an operation accomplished with the exclusion operator. As a simple example of the use of operators, consider the creation of a launch vehicle. The augmentation operator is used to add strap-on boosters for heavy lift capability, while the substitution operator could be used to express the change of a launch fairing.

Return to Table of Contents

2.4.3 Principles Supporting Extensibility

Four key principles support extensibility- modularity, ideality/simplicity, independence, and integrability. These principles were originally linked to “flexibility” by Schulz and Fricke (1999) and are briefly summarized below.

2.4.3.1 Modularity

The first principle supporting extensibility is modularity, defined by Baldwin and Clark (2000) as:

“A module is a unit whose structural elements are powerfully connected among themselves and relatively weakly connected to elements in other units. Clearly there are degrees of connection, thus there are gradations of modularity.” (p. 63)

The principle of modularity enables complex problems to be broken down through a hierarchical structure. Changes internal to a module are isolated at the module boundaries, limiting the cascading impacts of a required change. Expense is reduced in development, test, hardware exchange, etc. Changes made to a modular system can be described in terms of the modular operators described in the previous section.

2.4.3.2 Ideality/Simplicity

Ideality is defined by Schulz and Fricke (1999) as the ratio between useful and harmful/undesired effects, a notion of design efficiency (pp. 1.A.2-4, as an additional reference, see Suh, 2001.) This principle highlights the importance of the ongoing culling of unneeded functionality as a system evolves over time. Failure to do so increases system complexity unnecessarily, eventually making total replacement of the system a more effective option than change.

2.4.3.3 Independence

The independence axiom derives from the independence axiom in axiomatic design (Suh, 2001). Each function is satisfied by a different design parameter. Creating a decoupled design, in terms of functionality, produces a design that is more easily managed over time.

2.4.3.4 Integrability

Integrability relates to the degree to which a system’s interfaces are open, or flexible. Compatibility between elements is a critical enabler of flexibility. As an example, consider a docking interface on the space station. This interface would ideally be common across all future spacecraft, ensuring full compatibility. As an additional example outside the aerospace industry, consider the USB interface standard now used by many electronic peripheral devices such as keyboards, computer mice, flash memory cards, etc. The use of dedicated interfaces for each one of these devices would be highly inefficient, especially given the fact that only a small subset of the devices is needed at any one time.

2.4.4 Extensibility Summary

The concept of extensibility is critical to the creation of a sustainable exploration system. Extensibility must be an integral part of the exploration strategy to ensure that forward progress serves as a continually growing exploration foundation, even in light of policy direction changes. The concepts of extensibility are woven into the baseline missions and example conceptual designs within this report.

2.5 Historical Comparison: Antarctic Exploration

The history of Antarctic exploration provides valuable lessons for space system designers. From its inception Antarctic exploration and science shared many attributes and constraints with current space activities. Both, for example, have been highly dependant upon technological advances, including the need for complex logistics and cutting-edge life-support capabilities. Months of isolation during Antarctic expeditions present psychological hardships similar to those anticipated in extended Moon and Mars missions. More generally, Antarctic exploration, like space activities, has brought science into close involvement with politics. The following section examines how these factors affected some aspects of the development of Antarctic exploration and science, and draws lessons for space exploration programs.

2.5.1 Technology and Logistics:

“More than any other, Antarctic science is dependant on logistics, on the ability to place and maintain a scientist and his equipment in the right place at the right time. Expeditions to Antarctica up to 1925 depended on techniques of transport, communication, survival, which remained largely unchanged for 100 years…. after 1925 the development of mechanized transport, the airplane, radio and technology based on better understanding of human physiology, were to make access to the Antarctic, travel within it and survival in its hostile environment, much less difficult.” (Beck 1986, p.131).[ESS2]

The above quote summarizes well the disjointed nature of Antarctic exploration. Rather than follow a steady, continuous path of progress, the pace of discovery on the continent advanced through steps and jumps. Importantly, these advances in capability often resulted from the congruence multiple technologies, rather than any single technical development. Each jump offered great advances in knowledge returned per expedition, a situation that should be anticipated and exploited in the design of space exploration systems.

Most significant of these advances involves a shift from what has been termed the “Heroic” age to the Modern age of Antarctic exploration. The Heroic age is roughly delineated as the period from 1895 to the dawn of the First World War in 1915 (Walton, 1987). It marked a dramatic shift in capability from the previous era because of the use of liquid fuel, however, due to the still rather primitive methods of transport and “life support,” expeditions during this period often brought extreme hardships. National prestige, sovereignty, and personal fame—not science—motivated exploration during this period.

The Modern age begins roughly with the American expedition lead by Richard Evelyn Byrd from 1928-1930. It is characterized by the comprehensive use of airplane travel, electric communication, mechanized transport, and thus continuous logistical support (Fogg, 1992). Most of these technologies had existed for some time, and had been tested and refined through previous expeditions. Byrd’s expedition was the first to coordinate them systematically, increasing the amount of data collected by orders of magnitude. The following table summarizes the major technical advances that enabled this shift, as well as the impact on exploration capability and knowledge return. Systematic use is defined as use in everyday operations, as opposed to sporadic use and testing.

Technology	Introduction for Exploration	Systematic Use	Mission/Logistics Impact	Initial Knowledge Return Impact	Space-based equivalent
Radio Communication	1911	1929 (Byrd)	Coordination, safety	Immediate news of success increased public interest	Satellite Communications
Combustion Engine (land travel)	1907 (Shackleton)	1933 (Byrd)	Outdoor activity and travel in harsher conditions	Distribution of heavy seismic equipment	Rover
Airplane	1929	1928 (Byrd)	Pre-positioning for extended expeditions; Aerial photography	1 field season of land-based observation per hour (4000 square miles)	UAV's, Pre-positioning technology
Ice Breakers	---	post-WWII	increased access, extended access	More feasible permanent base	cyclers

Implications can be drawn from these examples for space exploration. Advances fall into rough classes of technologies with analogues in space systems. Combustion engines, which enabled the equivalent of surface rovers, had a great impact on the kinds of fieldwork that could be executed. Their introduction created the possibility of distributed use of heavy equipment for seismic operations. Their impact on mission logistics, however, was minimal at first.

The airplane and the radio had dramatic affects on knowledge return and mission logistics. The Byrd expedition was the first to fly over the pole. In doing so, he took over 1600 pictures covering 150,000 square miles, or the equivalent of 37.5 field seasons worth of observations using previous methods (Walton, 1984). He also discovered two Mountain ranges. The airplane also allowed for the possibility of pre-positioning and logistical support for inland bases.

Soon after flying over the pole, Byrd was able to communicate the accomplishment. His successful flight was beamed via radio immediately back to the United States, and this greatly increased US interest in Antarctica (Fogg, 1992).

An interesting feature of the progression of technological development is the lag between testing and systematic use. Radio communication and the combustion engine were tested with little impact in many expeditions before the Byrd expedition.

Interestingly, life support capabilities advance much more gradually than logistics technology. Man learned to live the extreme environment gradually, over several hundred years, with advances coming more through trial and error than scientific or technological breakthrough. (Fogg, 1992)

In many ways, NASA’s current task is to transition space activities from a heroic to a modern age. While national prestige and public attention will continue to play important roles in space activities, the time has come for more systematic and knowledge return. The history of Antarctic exploration demonstrates that when this occurs, as in the case with the first Byrd expedition, public attention and government funding are likely to increase rather that decrease. The next section examines this dynamic of science and politics.

Return to Table of Contents

2.5.2 Politics and Technology

Antarctic exploration requires support at the national level. Thus, as one author notes, “Antarctic scientists have often been used as political instruments and it would be unrealistic for them to think that their work can be isolated from the spheres of interest of economics, law, and [ESS3] politics.”(Klein 2000, p.319) The motivations behind various stages of Antarctic exploration are extraordinary in their similarity to space activities. They include included: prestige of geographical discovery, information and experience for navigation and commerce, and sovereignty. While science always played an important role during expeditions, and is now the single most important product of exploration, it is important to note that the underlying motivation for countries to invest in Antarctic travel has almost always been the “maximization of influence” rather than knowledge (Lee, personal communication).

Territorial issues became increasingly important at the transition from the Heroic to Modern age of Antarctic exploration. From 1908 until the signing of the Antarctic treaty in 1961, international tension rose and fell as countries made varied and conflicting claims to sovereignty. The following events in particular were important to this dynamic.

1908 and again in 1907 Britain issued formal territorial claim

1923 British claim the Roth Dependency

1924 French claim Adelie land

1933 Australia makes claim

~1939 Norway claims Dronning Maud Land

While the motivations behind these claims were complex and interrelated, the World Wars and advances in technological capability were certainly central factors. As with space activities during the Apollo Era, international interest, enabled by technological advances, fueled funding for exploration.

Byrd’s expeditions are a particularly interesting example of this kind of feedback loop in the US. As mentioned above, Byrd was the first to systematically incorporate modern logistical technology in his 1928 expedition. This mission and second following it were funded privately. Their success captured the public attention, increasing US popular interest in Antarctic exploration. (Fogg, 1984). At the same time increasing territorial claims and impending war on the European sharpened political and military perception of the strategic value of access Antarctica. The result was that Byrd’s third expedition, in 1939, was funded publicly and had the attention of President Roosevelt himself. In a letter to Byrd in 1939, Roosevelt explicitly stated the confluence of interests that lead to public funding:

“The most important thing is to prove (a) that human beings can permanently occupy a portion of Continent winter and summer; (b) that it is well worth a small annual appropriation to maintain such permanent bases because of their growing value for four purposes—national defense of the Western Hemisphere, radio, meteorology, and minerals. Each of these is of approximately equal importance as far as we know.” (Fogg, 1984, p.162)

Following the Second World War, international interest in Antarctica increased together with improved access. Antarctica exploration was facilitated by the use of ships designed specifically for working in ice, including modern ice-breakers. (Walton, 1987) In the tense environment of the Cold War, the ability to access Antarctica, much as with space, was itself justification for doing so. As is often the case, science was the veil behind which these interests developed. One state department official, Henry Dater, makes clear how these issues were interrelated in a letter he wrote in 1959:

“Because of its position of leadership in the Free World, it is evident that the United States could not now withdraw from the Antarctic…national prestige has been committed…. our capacity for sustaining and leading an international endeavor there that will benefit all mankind is being watched not only by those nations with us in the Antarctic but also by noncommitted nations everywhere. Antarctic simply cannot be separated from the global matrix. Science is the shield behind which these activities are carried out.” (Beck, 1986 p. 64)

While this view is a product of the geopolitical context, it illustrates how various factors can coalesce to form a sustainable program from a political perspective. The Byrd expeditions from before WWII had demonstrated American technical superiority in exploration and proven that modern technologies could be used to improve access to the continent. After the war, politicians and diplomats began to view exploration as an important symbol for global cooperation and competition, and were committed to continuing operations. Once implicated, national prestige and technical capability became intermingled, heightening the perception of value of continuing exploration.

Conclusions – Exploration and Sustainability

An important lesson that the history of Antarctic exploration provides for space exploration system designers involves the interplay between news, knowledge, technology, and funding. While Arctic exploration progressed slowly for decades, it was marked by distinct stages of increasing capability and increased interest. As the Byrd expedition illustrates, quite often advances in logistical and knowledge acquisition and transfer capability translate to increased political interest and funding. The spread of news creates public interest, while increased knowledge and logistical capability creates military interest. Both can generate funding for further expeditions, thus creating a positive feedback loop of discovery and technological development. Figure 6 illustrates the salient aspects of the feedback loop, which enabled the Byrd expeditions.

Figure 6: Positive feedback loop for exploration

Of course the real dynamics behind such a process are complex and varied. Byrd’s expedition occurred at a time when international interest in Antarctica was increasing for many reasons. Still, these reasons are at least enabled, if not intimately connected with increasing logistical capability and knowledge creation. Such dynamics are worth investigating for the sake of creating successful exploration systems in the future.

Return to Table of Contents

2.6 Designing for Sustainability: A Process

MIT’s 2004 spring class in Space Systems Design investigated the design of extensible space system architectures. A central difficulty in this task was the shear complexity of the problem, and the lack of an established methodology to design system architectures. An important result of the investigations was thus the methods developed to approach the problem, and the process by which “sustainability” could become central to design decisions. The end result was an iterative and holistic approach to the problem, which will hopefully inform future space systems architecture projects.

It should be stressed that not every aspect of the process described was completed rigorously during the semester. Rather, the process represents a way to integrate the lessons learned and eventually create a systematic architectural design. Of course every element of this process did not proceed in clear and neat steps. Most of the steps were iterative within themselves, and individual elements were re-worked as

The underlying goal of the design process was to develop an integrated strategy that could quantify how the system reacted to changes in the environment. Rather than create a point design to accomplish a Moon or Mars expedition, the class wanted to demonstrate that various scenarios could be anticipated and addressed during conceptual design and, as importantly, that the elements designed to address these scenarios (which would likely make the system sub-optimal from a point-design perspective) could be justified quantitatively. A strategy includes various scales of Moon and Mars missions, robotic scout missions, and considers the program changes such as budget cuts and regulatory constraints.

Figure 7 illustrates the five step process arrived at to create the strategy. An important goal was the establishment of common operations and across manned Moon, Mars and potentially asteroid missions, as well as through stages of missions at each body. Common elements defined baseline architecture forms and operations, from which options could be created to address specific missions and changing scenarios.

Figure 7: Space systems design process

The first three steps in the process identify common forms and functions needed to explore the Moon, Mars and other destinations. Two teams conceived of staged Moon and Mars missions, and created matrices with functional requirements for each stage. With these functional requirements, a simple Venn diagram captures the relationship of requirements between the Moon and Mars. An interesting feature of this part of the process involves the ability to identify how formal elements can be extracted from functional requirements based on commonality between Moon and Mars needs at various levels. “Options” can be created to supplement the core needs, based on requirements outside of the intersection of the circles.

Functional Commonality Mapping thus revises the forms created to enable various Missions. The two teams must then return to the mission storylines and establish how and whether mission objectives can still be met with the revised forms, and alter staged missions accordingly. This iterative process can continue until a satisfactory level of refinement is achieved.

It was found that this iterative part of the process reveals key trades that need to be made with respect to commonality and architecture operations. Based on our designs, trades on issues such lander design, rover design, aerobraking capability, and operational capability processes such as the use of the Earth-Moon Lagrangian points, could not be solved by commonality mapping alone. The next step of the process is thus to evaluate the key trades revealed by the first three steps of the process.

In order to create a flexible strategy, however, it was important to evaluate these trades with consideration for the value of flexibility and robustness, not just optimality. Tool such as real-options, multi-attribute utility theory, and decision analysis, can be used to carry out the trades while preserving system flexibility, thus creating a rigorous development strategy and architecture.

Chapter 6 addresses how these tools can be used to evaluate strategic and technical options. The strategy includes staged deployment of Moon and Mars missions, with development options forming branches from the baseline mission. Ideally the aspects of the system designed early in the strategy will minimize the need for redesign if new directions in the strategy are taken.

As noted, the full strategy was not generated during this design course. Instead, various aspects of the process were addressed and tools were conceived to facilitate their design in later studies.

Return to Table of Contents

3. Knowledge Delivery: The Core of Exploration

3.1 Explanation of the view

An extensible space exploration infrastructure may be modeled as a mass transportation system, but also as a knowledge delivery system, since mankind is sending robotic and human explorers to space for the purpose of exploring and returning knowledge about the Moon, Mars and Beyond.

To justify knowledge as the deliverable to the stakeholders one must investigate why knowledge is the deliverable and who the stakeholders are. To answer the first question, one must first understand why do humans explore. To summarize, the three main reasons are

To expand the knowledge of our surroundings
To improve the technological leadership of the United States
To inspire interest in science and technology

Knowledge is the product of the exploration process. The knowledge of our surroundings is closely tied to science. Technological leadership is knowledge delivered to the technologist and explorers. The third point is that inspiration in science and technology is the knowledge delivered to public and commercial enterprises. In other words, the knowledge gained by the space exploration system is the value-added delivery to the beneficiaries or stakeholders. Therefore, to ensure the maximum value delivery, one may model the space infrastructure as a knowledge delivery system. Knowledge returned may be categorized as scientific knowledge, resource related knowledge, technical knowledge, and planning related knowledge. To build up the argument, first one must understand the value delivery to the scientists, which is diagramed in Figure 8. To understand the value identification, the goal of the space infrastructure is to increase the quantity and depth of scientific knowledge of the solar system by sustainably and successfully exploring the solar system, specifically the Earth, Moon, Mars, and Asteroids (EMMA) using an affordable and extensible human and robotic exploration system for the immediate benefit of the scientific community.

Figure 8: Value delivery to scientists diagram

The value delivered to the technologists and explorers is an increase in the quantity and depth of resource and planning related knowledge of the solar system by sustainably and successfully exploring the solar system, specifically the EMMA using an affordable and extensible human and robotic exploration system, and the previously gained resource. The value delivery can also be seen in Figure 9.

Figure 9: Value delivery to technologist/explorers diagram

In addition to the scientists and technologists/explorers, knowledge may be returned for the benefit of the United States public and mankind. NASA and the US government, international partners and commercial enterprises may derive additional knowledge benefit. The full objective process methodology map is shown in Figure 10.

Figure