Seed Factories/Notes

The earlier parts of this book introduce the idea of Seed Factories, their history, component concepts and methods, and design process. This is followed by several design examples. The intent is a logical flow in the discussion for the reader's understanding. Developing the Seed Factory concept is a work in progress, so there is material that is either too new or not yet organized into the main text of the book. Developing the design examples in their full detail also generates a lot of material that would interrupt the flow of the main discussion. We therefore place place both of these kinds of material here in the Notes section. In effect they form a kind of "engineer's notebook". Historically these were paper notebooks and file folders that each designer kept for their work. The more modern approach is to use a shared electronic project database, so that all the current data is available to all the designers. So another reason to include these notes in the Wikibook is as an example of a collaborative project database. The Wikimedia format can only handle text and some file types. Other file types, such as spreadsheets, software code, and editable drawings will be stored elsewhere and linked.

General Notes
These general notes refer to the book as a whole, or the discussion in sections 1.0 through 4.0. This is followed by notes for the particular design examples.

Notes Background
These notes originated from the later parts of the Wikibook Space Transport and Engineering Methods. That book envisions a program to improve life on Earth by the use of advanced technology, and then enable inhabiting more difficult environments, eventually beyond the Earth and into space. To relocate to and operate in a difficult environment imposes a cost to deliver equipment and supplies. The more difficult and distant the location, the higher the cost. You can calculate that cost as (cost/kg for delivery) x (mass delivered). Reducing either factor will reduce the cost, but for space locations the most attention has gone to the delivery cost, in the form of designing better rockets and propulsion. The mass to be delivered can be reduced by methods like using local resources to make equipment and supplies, and by recycling materials on location.

The Seed Factory approach, sending a starter kit which self-expands to a larger production capacity, had been previously proposed as a way to use the local resources in space without having to launch massive amounts of industrial equipment. This author (Eder) first realized the same approach could be applied to building the Earth production capacity and launch site to reach orbit. That led to the idea of using it on Earth for general production, not just for space applications. Delivery cost is not as big a factor on Earth, but cost of production is. By reducing the amount of equipment you need to buy for a factory, using cheaper raw materials, supplying it's own power to operate, and high levels of automation, those costs could be greatly reduced. Since Earth is a more immediate and much larger market than space, the work shifted to how to develop terrestrial Seed Factories. That work became extensive enough that a separate book (this one) seemed justified. The accumulated study and design notes were shifted in bulk to here, and re-organizing them into the appropriate sections of the current book was started. That is still in progress, so the following sections are fragmentary, repetitive, and may not match content elsewhere in the book.

Study Goals and Plan
The notes here represent a concept study in progress, and not a final report. The sections show the work as it is developed in their full detail. This is for a couple of reasons. First, as a tutorial for others, we feel it is important to show all the steps in the process. Second, the authors have to record their work somewhere in the process of doing it, and this is it. As a "work in progress", the study will exhibit gaps, inconsistencies, and repetitiveness until complete. Normally these deficiencies are cleaned up in producing a final report or technical paper in order to present a clear summary of the analysis. In this case, the clean up is by organizing the notes into the main part of the book for a textbook-style discussion of the concepts, and keep detailed design and calculation data for specific examples here in the notes. That keeps them available for reference without interrupting the discussion.

The original Seed Factory study work identified several candidate technologies, and questions about to how best to use them, and whether they are enough of an advantage over conventional methods. The goal was to develop the ideas for these technologies in enough detail to start answering the questions.

The initial candidate technologies include:


 * Self-Expanding Production - producing increasing amounts of items internal to program locations, including more production equipment. This reduces what is needed to set up a new location to a starter set, reducing costs. Growth is by adding new equipment types, making larger versions of existing equipment, and by direct copies of the starter set.
 * Modular Design - Using standardized sizes and interfaces so that production elements can be added or rearranged without custom work.
 * Generalized Resources - Using widely distributed materials and energy as sources, rather than requiring high grade sources only found in certain places.
 * Cyclic Flows - repeat use of materials by recycling, repair, and closed-loop processes. This reduces the need for new materials and amount of waste products.
 * Distributed Operations - use of remote control, robotics, and automation to decouple where the people are and where the production elements are. This allows each to be in their optimum locations, rather than brought together in one place because there is no alternative.  The more difficult the location, the more desirable it is to start construction with fewer people, and the more important remote/robotic/automated methods are.

The questions to be addressed include:


 * What production functions are needed in various environments, starting with the easiest?
 * What should the starter set of elements contain, vs items added later?
 * How well do existing technologies meet desired goals, and what new technology needs to be developed?
 * How should production output be divided between adding more equipment, products for use in the program, and items for outside delivery to cover costs?
 * In expanding capacity, do you add more copies of existing equipment, build larger versions, or add new elements that can do new processes?
 * What is the right mix of direct human labor, remote operation, robotic, and automated tasks, and how does that evolve?
 * How much advantage do these advanced production methods have over existing/conventional methods?

 Study Process 

The Systems Engineering methodology was chosen to perform the study. Seed Factories are likely to be a complex design, and Systems Engineering is suited to complex projects. The goal for the study was to answer the questions noted above. It was envisioned in the context of a larger program to improve the quality of life and expand civilization to new locations, and inherited some goals and requirements from that larger context. To for the purpose of the study, we assumed a concept for the program as a whole, and then concentrated on the initial development of a single location Study steps include:


 * Requirements Analysis: Pass down such requirements and evaluation criteria from the program as apply, and add new ones if needed.
 * Functional Analysis: Break down production into sub-functions according to location types, to see if there are unique ones,
 * Requirements Allocation: Distribute our requirements to each lower level function.
 * Modeling and Define Alternatives: Developing a production model and identify a set of alternative approaches to perform the various functions.
 * Optimization and Selection: Develop the alternatives in enough detail to use within the model, then optimize parameters and select the best options,
 * Synthesis and Baseline: Combine the best choices for the various functions into a coherent whole and document a concept baseline to use for further work.

The resulting concepts are then assessed against the original questions to see how well they perform. The study remains remains incomplete until other parts of the program are more developed.

Initial Seed Factories Discussion
Improved technology is one of the main ways that civilization has advanced. Before starting the technical analysis, we describe the candidate technologies and the motivations for using them.

Note: The discussion that was here has been moved to sections 3.0, 7.0, and 8.0.

Program Concept
In the Space Systems book from which these notes came, an overall program for expanding human civilization is envisioned. A particular temperate location project was included as the first step in that program. In this book we separate that first location into the Community Factory example, which can stand on it's own, and treat the larger program as a separate example for Remote Locations. We inserted two new examples between these for industrial and distributed factories, to better explore applications for the Seed Factory concept. The remaining notes not yet incorporated follow:

The Program Concept describes in words what the whole program is intended to do, and how it will be operated and maintained. The Location Concept does this for the particular location type in this study. These are the starting concepts, which may get updated as the study progresses.

Temperate Location Concept
The first location established after Phase Zero (Technology Development) is at a Temperate location, where temperate is defined as an environment where the middle 90% of people currently live, with 5% at each extreme of conditions. A "location" does not have to be on one physically connected piece of land, but it should have all it's elements close enough that they can easily operate with each other. This location may carry over prototype elements from Phase 0, or use the prototype equipment to build the first full scale elements. The goal for the first location is to supply most of the long term needs for 660 people, building up to this at a rate of 75 more people per year. This can be considered a form of real estate development, except it includes living, working, and transportation elements designed to work together.

Habitation - The initial set of people and equipment may be distributed in the existing local community if the site development is starting from scratch. The future residents are assumed to self-build most of the location, since they will be the main future owners and operators once it is built. As the habitation portion of the location gets built, they can move themselves there.

Production - The output of the starter set (seed factory) is split among making more equipment for expanding production, satisfying the human needs of the owners, such as building materials and food, and external production for sale or trade. The question is who decides how much of each category to produce, and what specific items to make? That can be sales-driven, where outside buyers determine what gets made first, and leftover capacity gets used for growth and owner's needs; owner-driven, where the factory owners or designated managers decide what to make; or contributor-driven, where contributors of capital and labor can decide individually what to make according to their share of the contributions.

Operations - Modern computers and networking make it possible for the people to work remotely from the equipment. We do not require it be done this way, just have it available as an option. Distributed production and staff can save on daily commuting costs and travel time, and let equipment be in optimized locations, rather than centralized like a traditional factory. This forces higher communications cost, but that component is rapidly getting cheaper. Moving people less and distributing the physical production may result in more transportation for physical items. Another aspect of operations is who is controlling things. This can be humans locally, humans remotely, and computers directly (automation and robotics) or remotely. We expect to have a shifting mix of all of the above mentioned options, with a trend towards more automated and less human work. One incentive to develop more remote and automated methods than are strictly needed for the first location is that future more difficult and hostile locations would be more expensive and less desirable for humans. Another is that change of job or work location becomes decoupled from the preferred location to live.

Maintenance - The production part of the location makes most of the products needed to build out the full site. Therefore our maintenance concept is to use the same production capacity to repair, replace, or modify the location elements as needed. Any items that cannot be repaired are returned to production as scrap, or exported as waste, but by design we try to minimize this.

Requirements Flowdown
Note: These are requirements extracted from the larger Human Expansion program and passed down to the first location. They have been redone for the Community Factory example in Section 5.0 in this book.

This step passes down requirements from the overall program level to lower tier elements. The existing baseline (human civilization in its present state) has an approximate score of 75 points, so the goal for Phase I of the program is to do 10 points better, or 85 points. Setting this score level generates a set of parameters to apply for the first location in Phase I to use as design requirements. Some of the requirement values will be quite challenging to reach, but by setting difficult goals we will identify what areas of technology need improvement and guide future research. We also expect to see some requirements adjusted up or down when they turn out to be easier or harder to reach than expected. The initial values are:


 * First location design increment = 75 people, with growth at 75 people per year.
 * Growth capacity after start-up = 11% per year
 * Local matter and energy resource use = 85%
 * Self-production by economic value = 85%
 * Recycled mass flow excluding fuel = 85%
 * Reduction in labor-hours from automation and other technologies = 85%
 * Local autonomy of labor and operations = 85%
 * Equivalent GDP/person = $156,000 (Jan 2013 US $ base for all costs)
 * Long term ratio of materials & energy production to internal needs = 10.5
 * Technology development cost = $66 million for temperate location.
 * Location construction cost = $76,000 per person.
 * Technical uncertainty = 7.5%

We can then further allocate requirements for the production function within the location:


 * After start-up, it should be able to meet 85% of the basic needs (shelter, food, utilities) of 75 people.
 * After start-up, the production function should have produced 85% of the total value of the location, including itself. Sales of surplus products is included in calculating net value.
 * The production function should be able to expand itself by 11% per year (with some added parts from outside).
 * The production function should be able to recycle 85% of location mass flows, not counting fuel.
 * Labor hours/unit output should be 85% below 2013 industrial production average.
 * After start-up, at least 85% of labor and operations control will be internal to the location.
 * Long term ratio of materials and energy production to internal needs is 10.5 to 1.
 * Development of production technology to be less than $66 million less habitation and transport development.
 * Net cost of the production elements should be below $5.7 million for a location supporting 75 people.
 * A technical performance uncertainty of 5-10% for the production function, balanced with habitation and transport uncertainties.

From the location type and size we can also establish:


 * The design must function in a temperate climate near a developed area.
 * Local operation should require no more than 85% of time to supply 85% of needs x 50% labor force participation x 75 people = 32 full time staff.

From higher level program requirements we can also set:


 * To encourage participation, the design of the prototype should use open-source methods, while ownership of the hardware and outputs will belong to the project contributors.

These requirements then become goals for the technology development phase to meet.

Evaluation Criteria
We need a way to optimize the design and choose among alternative concepts. The standard Systems Engineering approach is to define various parameters for what is "better", such as cost and growth rate. These parameters are converted to a common measuring scale, and whichever choice gives the highest score on this scale is the one selected. The measures are arbitrary, since they are based on customer desires, but they are a way to model those desires in a mathematical way so that design choices can be made. If one changes the desires, then the technical design would also change. The customer here is assumed to be a non-profit foundation building the prototypes for research and the benefit of people generally. Measures were previously defined at the program level (Section 5.1, page 4), so to be consistent we will use the subset of these measures which apply to the production function, keeping the same formulas and relative weights for scoring purposes.

At the moment, we are developing a concept for just the first temperate location, so the full set of evaluation criteria are modified as follows:


 * 1.2 Program Scale - The initial size for the first location is fixed at 75 people, with growth to 660. This will be a fixed value for the present work, so is dropped as a point of comparison or optimization.  At least one performance parameter must be fixed so the remainder can be optimized against it, and we choose the number of people supported at the location to be the fixed one.
 * 2.1 Number of Locations - For the first location, this is fixed at one location in a temperate environment, so it is also dropped from comparisons.
 * 2.2 Growth - We keep this parameter as a variable of the design options, with a goal of 11%.
 * 2.3 Improved Technology - We keep all five of these parameters, with a goal of 85% each.
 * 2.4 Quality of Life - We keep this parameter, with a goal of $156,000 per capita GDP equivalent.
 * 2.6 Resources - We keep this parameter, with a goal of (materials and energy production)/(internal needs) = 10.5.
 * 4.1 Total Development Cost - Only the Earth component is considered here, since the single location is on Earth. We keep this as a key parameter, with a goal of $890,000 per person x 75 people = $66 million.  This is a peak number net of outside sales.
 * 4.2 New Location Cost - Again, only the Earth component is considered, and we keep it as a key parameter. The goal is $76,000 per person x 75 people = $5.7 million for the first location.
 * 4.3 Earth Launch Cost - We skip this parameter since this location is on Earth.
 * 5.1 Technical Risk - We keep this as an important parameter during development, with a goal of 5-10% once prototype hardware has been demonstrated.
 * 6.1 New Location Risk - We keep this as a design parameter, with a goal at the end of technology development of 38% relative casualty risk.
 * 6.2 Population Risk - We keep this parameter, but since the size is only 0.05% of the total program, we only consider the nearest 3.5 million population for effects. The goal is a 17% reduction to general population risk, but this is likely to be hard to reach.
 * 7.1 Biosphere Security - We keep this parameter, but again reduce to goal to 0.05% x 178,000 = 89 species preserved.
 * 7.2 Survivability - We keep this parameter, but adjust the goal to 0.0085% compensation for critical risks.

We will consider an added parameter for production:


 * 2.3 Starter Complexity - Fewer different types of devices in the seed factory mean less item types to design for and maintain.  Since products in general are made of at least several different materials, we will assume a set of 5 starter machines would get a score of 100%. A survey of production methods shows there about 220 physical fabrication processes, and 60 chemical unit operations and reaction types.  We prefer not to have anything close to all 280 processes in a seed factory, so give a score of 0% to one with 60 starter items.  We interpolate via the formula score = 160% - (40% * ln(NP)), where NP is the number of processes in the seed factory.  Thus the worst case of a seed that needed every known process would get a score of -65%.  We will use a relative weight of 1 point, along with the other technology parameters.

For consistency we keep the same weights as the program level measures, which results in a nominal maximum score of 79 points, and a design goal of 85% of that, or 67 points.

Functional Analysis


For this analysis we consider the location as a whole as the system, and draw a boundary around it to separate it from the outside environment (Figure 5.3-1). Input and output flows that cross the system boundary come from the physical environment and the remainder of human civilization, which includes technology developed in Phase 0. Within this system boundary, we have previously identified three top level functions of production, habitation, and transport. There may be other functions at this level not identified yet. We do not show the flows between the functions or to and from the system boundary because that would make the diagram too complicated. Instead we use a spreadsheet which tabulates each flow and box on the diagram as rows, and the various resources that make up that item as columns. The Input-Output Model on our Desktop Space Program project on Sourceforge is a generic version of such a spreadsheet, and the Temperate Location Model is the particular one being developed for this study. The design must meet the rule that total flows and functions sum to zero, where outputs have positive values and inputs have negative values. Otherwise something is appearing or vanishing without explanation. Enforcing this rule ensures that all necessary items are identified and accounted for.







Next we break the top level functions into their next lower level parts. We have done this in what seems a natural way by type of function. Part of functional analysis is to identify what needs to be done, but not exclude possible design solutions for how it will be done. Therefore the functions are named in terms of a task and not the method to implement the task. Figures 5.3-2 to 5.3-4 show this next-level breakdown. Again, we have not tried to show the flows between functions because there would be too many of them in the illustration. It should be understood that each box will have multiple flows entering and leaving, and the details will be tracked in our spreadsheet. There can be more than one way to break down a function, leading to different designs to satisfy it, but we will try this one first before looking at alternate ways. Since this study is particularly interested in the production function, we may do additional breakdown of details below what is shown in Figure 5.3-2.

Function Descriptions
In addition to the diagrams, we also write descriptions of the functions to further explain what they do. They are numbered according to the more general Human Expansion project scheme to be consistent with later work.


 * F.2.1.1.1 Provide Production Capacity

(Moved to Section 5.2)


 * F.2.1.1.2 Provide Habitation Capacity

(Moved to Section 5.2)


 * F.2.1.1.3 Provide Transport Capacity

(Moved to Section 5.2)

Allocate Requirements to Functions:
We now take the requirements listed above under Requirements Analysis, and parcel them out to the lower level functions to create the smaller subsets that each lower function is responsible for meeting. All of these requirements are to be fully met at the end of the growth of the location, and met as well as practical before then.


 * Design Increment - Provide support for 75 people initially, and adding 75 more per year to reach 660 total at first location. Assigned to F.2.1.1.1, .2, and .3 as "Provide [Production, Habitation, Transport] Capacity for 75 people per year" respectively.
 * Growth Capacity - Provide growth capacity after startup of 11%/year. Assigned only to F.2.1.1.1 Production as "Produce new equipment capable of 11% increased output per year."  Outside components will be needed for the new equipment, and growth is measured by economic value of the output.  Additions which take multiple years are distributed over the years for credit towards the goal.
 * Local Resource Use - Use 85% local matter and energy. Assigned only to F.2.1.1.1 Production as "Provide 85% of location matter and energy needs from local resources by economic value."
 * Self Production - Produce 85% of economic value locally. Divided between Production and Habitation/Transport functions depending how location labor decides to work.
 * Recycled Mass Flow - Re-use 85% of mass, not counting fuel. Assigned to all three major functions as a 15% limit on unused waste flows by mass. This likely will require high level of recycling of water and food and resulting wastes.
 * Automation - 85% reduction in labor hours relative to US average. Assigned to all three major functions, with most of the reduction in the production area.  Some gains in household maintenance and food preparation, and automated transport should be possible.
 * Autonomy - 85% of the control of needed operations at the location come from local sources (humans and control equipment).
 * Starter Complexity - Goal of 7 devices in the starter set of machines. This counts the major machine items, and not accessory heads, bits, tooling, or conventional small shop tools.
 * Quality of Life - Equivalent GDP/person of US$ 156,000 (Jan 2013 base). Assigned to a combination of  mostly Habitation and Production, and perhaps a little for transport.  Equivalent GDP means counting internally produced and used items at their market values, and adding external sales of products and services, including surplus labor not needed for internal operations.
 * Resources - Long term ratio of materials and energy production/internal needs of 10.5:1. Assigned to production function.  Surplus production goes to location growth or sold for income.
 * Development Cost - Maximum net development cost of $66 million for temperate location, divided between all three functions, but likely most will be under production. Net cost is after any sales during development.
 * Location Cost - $76,000 per person, divided between all three functions. This is the recurring cost of building the first location, after development.
 * Technical Risk - Performance and design uncertainty of 7.5% divided between 3 main functions. The goal by the end of development is to have reduced the uncertainty to this level.  It will be higher at the start of development.
 * Location Risk - Life and casualty risk of 38% relative to US average, divided between 3 main functions.
 * Population Risk - 17% reduction to population risk for nearest 3.5 million population. Self-reduction counts towards this, but general risk reduction may be difficult on a small scale.  Allocated to all three functions.
 * Biosphere Security - Preserve 89 species outside their normal environment range, either stored or living. Divided between production and habitation, depending on species.
 * Survivability - Provide 0.0085% compensation for civilization level critical risks. Assigned to production function.  Likely areas are resource depletion or carbon reduction, but this will be a difficult goal.

Collect Assigned Requirements by Function:
Note: Tables moved to Section 5.3