Content uploaded by John R. Olds
Author content
All content in this area was uploaded by John R. Olds on Jun 17, 2014
Content may be subject to copyright.
IAA-00-IAA.1.3.05
Space Tourism: Making it Work for Fun and
Profit
J. Olds
D. McCormick
A. Charania
L. Marcus
Space Systems Design Lab
Georgia Institute of Technology
Atlanta, GA
USA
51st International Astronautical Congress
2-6 Oct 2000/Rio de Janeiro, Brazil
For permission to copy or to republish, contact the International Astronautical Federation
3-5 Rue Mario-Nikis, 75015 Paris, Fance
IAA-00-IAA.1.3.05
1
Space Tourism: Making it Work for Fun and Profit
Dr. John R. Olds*, David McCormick†, Ashraf Charania† and Leland Marcus†
Space Systems Design Laboratory
School of Aerospace Engineering
Georgia Institute of Technology, Atlanta, GA 30332-0150
ABSTRACT
This paper summarizes the findings of a recent
study of space tourism markets and vehicles
conducted by the Space Systems Design
Laboratory at Georgia Tech under sponsorship of
the NASA Langley Research Center. The
purpose of the study was to investigate and
quantitatively model the driving economic
factors and launch vehicle characteristics that
affect businesses entering the space tourism
industry. If the growing public interest in space
tourism can be combined with an economically
sound business plan, the opportunity to create a
new and profitable era for space flight is
possible. This new era will be one in which
human space flight is routine and affordable for
many more people. The results of the current
study will hopefully serve as a guide to
commercial businesses wishing to enter this
potentially profitable emerging market.
NOMENCLATURE
AF airframe
DDT&E design, development, testing and
evaluation
FY2000 fiscal year 2000
IOC initial operating capability
IRR internal rate of return
LMNoP Launch Marketing for Normal People
NASA National Aeronautics and Space
Administration
NPV net present value
RLV reusable launch vehicle
SG&A Selling, General and Administration
TAT turn around time
TFU theoretical first unit
TIF time in flight
INTRODUCTION
Study Overview
The present research was conducted in four
distinct phases. Phase 1 consisted of the
development of a new flexible modeling tool for
simulating the future space tourism launch
market. This new tool, LMNoP, predicts the
number of passengers (space tourists) available
to the market in any given year as a function of
ticket price, expanding market size, perceived
reliability, number of launch sites, orbital vs.
sub-orbital capabilities, passenger
accommodations, airframe lifetime, and other
variables. Coupled with launch vehicle
characteristics such as development cost,
turnaround time, recurring cost, and number of
passengers, the LMNoP model allows an analyst
to model the economic attractiveness of any
proposed space tourism scenario. LMNoP is a
stochastic model and directly treats uncertainty
in market size and growth using Monte Carlo
simulation techniques. The economic results are
therefore distributions of expected return on
investment, net present value, etc. for an
optimized ticket pricing strategy. Phase 2 has
tested this new tool is tested on several proposed
space tourism transportation options to
determine if any makes a strong business case.
Phase 3 of the project has identified and
prioritized the major economic drivers for a
profitable business case and has useful
established goals/targets for the most important
* - Assistant Professor, School of Aerospace Engineering,
Senior Member AIAA.
† - Graduate Research Assistant, School of Aerospace
Engineering, Student Member AIAA.
Copyright ©2000 by John R. Olds and David J. McCormick.
Published by the American Institute of Aeronautics and
Astronautics, Inc., with permission. Released to IAF/IAA/AIAA
to publish in all forms.
IAA-00-IAA.1.3.05
2
vehicle characteristics (e.g. reliability > 0.9999,
investment cost < $1.5B). Phase 4 used the
sensitivities generated by Phase 3 to find an
economically viable space tourism transportation
option.
Background
As regular Space Shuttle and Soyuz flights make
spaceflight seem routine to many people, the
subject of private space tourism is making
appearances in the popular press with increasing
regularity.
Figure 1 – Space Tourism Theme Park.
The conclusions of many studies to date are that
this business area will be lucrative. Penn and
Lindley conclude that with near-term reusable
technology, a viable space tourism business can
be created using very high flight rates and
inexpensive propellants.1 They also conclude
that the market size is adequate to support the
industry. The argument was a the extremely high
flight rates, the cost of expensive cryogens
actually became a driving factor in cost, contrary
to current launch vehicles, where propellant costs
are small enough compared to other costs that
they can essentially be overlooked. To further
bolster reusable launch vehicle flight rates,
synergies between a high flight rate space
tourism model and a high flight rate cargo
market like space solar power were also
identified.2 A similar conclusion is reached by
Rogers who supports a shift in mindset for future
launch vehicle projects to vehicles with high
operability and low costs for launch.3
To assist the space tourism segment of the
industry, there are many other synergies with
ground-based industries such as theme parks and
advertising.4 These could help reduce some of
the economic burden when compared to an
exclusive passenger carrier activity. These
ground-based industries could also be enablers
for space tourism.
Factors such as this combined with the promises
of certain new technologies intended to make
human space flight both safer and more cost
effective, make private space flight seem more
likely than ever.
Motivation
Point - Spaceflight has intrigued the popular
consciousness since before mankind even knew
of its possibility. The vastness of the cosmos
combined with the feeling of discovery is an
experience enjoyed by most only vicariously
through astronauts. Just as atmospheric flight
was first only experienced by few onlookers
gawking at early barnstorming and select
members of the military, then progressed to be
experienced by only the very wealthy to the
current day or routine air travel, space travel
should eventually progress to the average person.
It is the destiny of spaceflight to follow this same
paradigm and open the heavens to the masses.
Counterpoint – That’s all great, but I want to
make money.
To date, it has been hard to get around
Counterpoint. Certainly, as evidenced by
government programs, it is technically feasible to
send humans into space for extended periods of
time and return them safely to earth. Thus, the
economic challengeis the only thing standing in
the way of the enjoyment of space for orders of
magnitude more people than enjoy it today.
What cost goals do the aerospace community
IAA-00-IAA.1.3.05
3
have to meet in order to bring this industry to
fruition?
To answer those questions as well as aid future
inquiries into the business of space tourism is
essential to its emergence. At the center of this
research is a stochastic cost analysis used to
evaluate several concepts, identify driving
factors in the economic viability in selected areas
of the design space and then use this information
in a cost-as-an-independent-variable analysis to
determine the “break points” for the values of the
input variables for the cost analysis. These
“break points” should show how far this industry
must go to be successful.
LAUNCH MARKETING FOR NORMAL
PEOPLE (LMNOP)
Overview
LMNoP is a new stochastic Microsoft Excel©
business simulation for space tourism created
during the course of this research. It takes
vehicle economic characteristics such as design,
development, testing and evaluation (DDT&E,)
theoretical first unit (TFU) cost, reliability, etc.
and inserts these data into a random process
simulation. This simulation then does a life cycle
cost analysis on the vehicle based on input from
a stochastic market demand model, a
consequence-based vehicle failure simulation
and a customer-appeal analysis module.
These then use pseudo-random number
generation to create a different scenario for each
recalculation of the model. The model is run on
the order of one thousand trials and a distribution
for economic evaluation parameters is generated.
These distributions provide economic feasibility
information in the form of probability
distributions.
Life Cycle Cost
LMNoP builds a vehicle development program
around projected space tourism market demand.
The financial qualities of that program are
determined from user defined programmatic and
cost variables. The company that is building the
vehicle is assumed to be the same as the provider
of launch service for the space tourists.
Figure 2 – LMNoP Economic Schematic
Program Definition
Economic
Financing
Schedule
Fleet Definition
Pricing
Reliability and Market Model
Mission and Costs
Equity, Cash Flows, and Depreciation
Debt
Financial Statements
Mission Spread
Non-Recurring Costs (Boosters + Propulsion)
Recurring Costs (Launch site + Failures)
Revenues (normal and add-ons)
Equity Calculation
Simple Net Cash Calculation
Depreciation
Annual Deferred Liability including:
Annual Financing Cost (Int. + Principal)
Maturity Date of Bonds
Principal Due At Maturity
Income Statement
Cash Flow Statement
Balance Sheet
Net Present Value Calculation
Economics / Financial Side of LMNoP
Summary Metrics
IRR
ROI
Revenue
NPV
Max. Exposure
Total Number of Failures
IAA-00-IAA.1.3.05
4
LMNoP does not have the capability to cost
concepts given a particular vehicle definition.
The costs in the model come from other sources
(such as from literature reviews for existing
concepts like the Soyuz or cost estimating
relationships for Hyperion10). These costs are
integrated into the LMNoP financial engine in
order to determine the full financial scope of the
project. LMNoP is robust enough to handle
different vehicle concepts, development
schemes, financing plans, and pricing structures.
LMNoP is also well suited to handle new
developments in operations through its use of a
site fee. A built in assumption is that no vehicle
will build its own indigenous launch facility
(with associated capital expenditures) but rather
pay user fees at some future spaceport or lease
operations at existing facilities.
The economic and financial portions of the
LMNoP model obtain inputs from the program
definition, flight reliability, and multipliers
section of the model. Financial metrics like
internal rate of return (IRR) and net present
value (NPV) are determined through calculation
of specific program costs. These are then
coupled with user-defined pricing with
associated multipliers that originate in other parts
of the model. Five sets of program definition
inputs are needed. These are broken into
economic, financing, schedule, fleet, and pricing.
Program Definition
The economic variables that need to be defined
for each analysis include the dollar year that all
subsequent values are based upon, inflation rate,
tax rate, discount rate, and average annual
interest rate (used for calculation of the interest
that needs to be paid on deferred liability or
debt).
The financing variables include those that
determine both the frequency and amount of
equity (i.e. stock) offered as well as the per-year
fixed and per-flight variable selling, general, and
administrative (SG&A) expense.
The scheduling variables include user
determination of initial operating capability
(IOC,) program termination, years for vehicle
development, and years to ramp up to full
operability. Before any flights can occur,
LMNoP (based upon user input) segments
airframe and engine development into
appropriate years before IOC.
The model can handle up to three new, separate
vehicle sub-developments in the program (with
the capability of modeling up to two stages for
each vehicle). This can account for the same
company building a sub-orbital vehicle and then
transitioning in a future year to an orbital
vehicle. For each stage of the vehicle (as well as
where appropriate its associated propulsion
module) the following fleet definition variables
are needed: passenger capacity per launch,
overall reliability, flight lifetime, turn-around
time, time in orbit, DDT&E cost, TFU cost,
learning effects, and government contribution
percentages.
The pricing variables include insurance
definitions, charges for failures, and site fee costs
per flight. Insurance in this case refers only to
vehicle liability insurance per flight based upon
the expected probability of failure (1- overall
reliability) multiplied by the TFU cost of the
vehicle’s airframe and engine. If there is a failure
in any particular year, two economic effects
instantly result: namely the company is out of
business for a specified number of years
(accepting a user defined one-time charge to
account for program recovery and victim
redress) and all subsequent insurance changes
per flight increase by a certain user defined
percentage.
If the vehicle is modeled as an already existing
development (i.e. like a Soyuz) a set recurring
cost per flight can be set. Yearly pricing options
include both static and varying (based upon
either a linear or quadratic pricing). Up to five
different revenue types can be used to account
for additional revenues from non-direct sources
(i.e. advertising on vehicle, television revenue,
etc.).
IAA-00-IAA.1.3.05
5
Financials
A separate mission and costs section determines
the spread of flights dependent upon market
captured for various prices. This translates into
non-recurring costs (booster/propulsion
development and government contribution),
recurring costs (launch site fees and business
failure charges), and revenues (from static/
variable pricing and revenue add-ons). Equity
calculations are then determined along with
associated deprecation schedules. Deprecation is
defined using U.S. government standards based
upon a 5-year depreciation of fixed assets. A
separate debt calculation is made with the
assumption that negative cash flows in any given
year (after accounting for revenue and equity
infusion) are paid off using either long or short-
term bonds (20, 15, 10, 5, or 1 year varieties).
For this financial analysis, the free cash flow is
defined in Eqn. 1.
(1)
All the above information is aggregated to obtain
the discounted cash flows and associated
summary metrics like NPV (for NPV, based
upon user defined discount rates).
Earnings before Interest and Taxes
(EBIT)
-
Taxes (tax shields from negative
income years carried over until
exhausted by tax liability)
-
Capital Expenditures (airframe and
engine acquisition)
+ Depreciation
= Free Cash Flow
Figure 3 – LMNoP Schematic.
Program Definition
Economic
Financing
Schedule
Fleet Definition
Pricing
Market Model
Price Regressed Data
Stochastic Reliability Multipliers
Comfort and Appeal Multipliers
Mission and Costs
Equity, Cash Flows and
Depreciation
Debt
Financial Statements
Economic Analysis
Summary Metrics
IRR
ROI
Revenue
NPV
Max. Exposure
IAA-00-IAA.1.3.05
6
Market Demand Model
The pre-adjusted market demand is based on a
literature search. This search focused on survey
results that specified launch market demand as a
function of ticket price. It resulted in two market
surveys that are used in LMNoP.
The primary source for market information is the
Commercial Space Transportation Study
conducted by a consortium of aerospace
companies for the National Aeronautics and
Space Administration (NASA.)5 This provides
information based on worldwide incomes and the
likelihood of those with sufficient income
interested in a space trip purchasing a ticket. This
represents a more bottom-up approach. The
second is a top-down approach by Nagatomo and
Collins.6 This provides market survey data to
augment the CSTS information. All market
information used is for worldwide demand.
Figure 4 – Market Curves for LMNoP.
This results in a population of results for each of
the price points of the investigation. To account
for this population spread, a normal distribution
is fitted to the data at selected price cross-
sections. From this, the model interpolates the
mean and variance of the normal distribution to
obtain the probability distribution for the number
of customers at a specified price. Then for each
year of the simulation, a random member of that
distribution is selected to be the number of
customers for that particular year. This results in
a randomly fluctuating customer base for each
simulation that tests the robustness of a project
against changing market conditions.
This market information is then fed to the
reliability and customer appeal modules for
adjustment before it is sent to the life cycle cost
model.
Reliability Module
The reliability module contributes to LMNoP by
placing a multiplier on the baseline customer
demand information provided by the market
module. When there are no failures, this
multiplier is unity and there is no change to the
remaining sections of LMNoP. Once a failure
occurs, the module begins to modify the market
demand as well as affect cash flow. Whether or
not a failure occurs is modeled by a constant
hazard rate for each year based on the number of
flights in that year. There is no break-in period or
age effects on reliability.
The most immediate impact of a failure in
LMNoP is a fixed charge to the operating
expenses of the company. This represents the
liability associated with carrying members of the
general public. This charge can be user-specified
and should be in line with the expenses
associated with an airline accident involving loss
of life. The one time charge should be punitive
enough so as to discourage reliability low
enough to cause failure.
The second aspect of a launch failure is a
complete shutdown of market demand and
therefore flight operations while the cause of the
failure is investigated and remedied. This period
of time can be more than a year and significantly
affects the profitability of a space tourism
concept.
The third impact of a failure is a slow linear
ramp-up in customer demand following a failure.
This is designed to simulate the rebuilding of
trust in the company over time after operating
successfully.
1000
10
9
1
10
100
# people
10
8
10
7
10
6
10
5
104
10
3
102
10
1
Nagatomo & Collins
CSTS Low
CSTS Medium
CSTS High
Ticket Price ($K)
IAA-00-IAA.1.3.05
7
The final impact of failure results from the
possibility of a second failure during the ramp-
up period. It is expected that this would
completely obliterate public confidence in the
project, driving market demand and therefore the
flight rate of the project to zero. In LMNoP, this
results complete business shutdown and halts life
cycle cost analysis.
Fig. 3 shows an example of the market multiplier
effect of a failure. There is a failure in year 25
and then another in year 30 during the recovery
period. This is fatal to the business and the
analysis of this case ends at that time.
Figure 5 – Consequences of Failure.
Customer Appeal Market Multipliers
It is obvious that certain entertainment value
factors of a space tour will increase desirability.
LMNoP divides these factors into comfort,
visibility, duration and availability.
Unfortunately, the literature search did not reveal
the quantitative effects of these intangible items
on customer demand, so engineering judgment
determined the values for each of these factors.
Comfort
Comfort is divided into four categories, all
directly modeled after airline comfort levels.
Comfort level for this model is primarily defines
by the amount of volume afforded each
passenger. LMNoP recognizes the following
categories of passenger comfort:
• Sub-Coach – This level of comfort is less
than that of the average Coach-level airline
flight. There is a minimal amount of room
with no amenities. This has a market
multiplication factor of 0.5.
• Coach – This level is the same as that for
airline coach class, with the exception of
food and beverage service. It is doubtful this
will be possible during an earth-to-orbit
ascent. This has a market multiplication
factor of 1.0
• Business Class – This offers more room than
coach, with the possibility of flight crew
service during extended flights. This has a
market multiplication factor of 1.5.
• First Class – This is everything a first class
passenger might expect on a major airline.
This has a market multiplication factor of
2.0.
Visibility
Visibility provides a better passenger experience
and affects the market model as follows:
• Multiple people per window – 0.5 times
standard market.
• One window per person – 1.0 times standard
market.
• One large window per person – 1.5 times
standard market.
• “Glass ceiling” view – 2.0 times standard
market
Duration
Duration of the flight also influences passenger
experience and therefore affects the market as:
• Sub-Orbital – 0.5 times standard market
• Single Earth Orbit – 1.0 times standard
market
• Multiple Earth Orbits – 1.5 times standard
market
0
0.2
0.4
0.6
0.8
1
1.2
1
4
7
10
13
16
19
22
25
28
31
34
37
40
43
Year
Multiplier
IAA-00-IAA.1.3.05
8
• Space Hotel – 2 times standard market
Availability
The number of global launch sites can affect the
market size for a space entertainment venture.
Here it is assumed that 3 launch sites enables
global market capture. This is based on the
assumptions of the market surveys that make up
the base global market model that the three main
markets for space tourism will be Europe, North
America and the Pacific Rim. A curve fit to the
market capture for 1, 2 and 3 sites was extracted
and this is used as a multiplier for the base
market model. This given in Eqn. 2:
(2)
CONCEPT RESULTS AND DISCUSSION
Overview
To both test the LMNoP model and see where
several concepts stand as far as their profitability
in a space tourism environment, LMNoP was run
on four concepts. They vary from currently
flying (Soyuz) to many years into the future
using a representative third generation launch
vehicle concept. All analyze the business case
for an owner/operator of some type of hardware
component for carrying people into space.
Soyuz Purchase
The Soyuz (Fig. 6) test is designed to test current
space tourism opportunities using the LMNoP
model.7,8 Because trips to Mir via Soyuz
capsules are already being marketed to an elite
clientele, this should give a relative idea of how
our modeling technique would evaluate such a
plan. The basic idea is to purchase a Soyuz flight
for a fixed price for 3 passengers from the
Russian government in exchange for an orbital
flight for paying passengers. This is a low up-
front investment space tourism strategy.
Figure 6– Soyuz Spacecraft and Launch
Vehicle.8
Concept Assumptions
Soyuz was selected to represent using a current
expendable launch vehicle in the space tourism
market. Because it used existing technology
DDT&E and TFU were assumed to be zero.
Also, because there was no risk associated with
developing a new launch system, the discount
rate for calculating NPV was chosen as 15%, the
lowest of all the candidate designs. The fee paid
to the Russian government is assumed to be
$28M.
Price Sweep
As is evident from Fig. 7, the optimal pricing
strategy is largely determined by the price paid
to the Russian government for the Soyuz launch.
This optimal price is very close to the maximum
of $10M per passenger for the LMNoP market
model. It is to be expected as the cost to the
space tour company is $9M per person on the
flight. This profit margin does not compare well
to the 15% discount rate. The price also means
this is not the gateway to space for the average
person.
sitesNumber _57735.0
IAA-00-IAA.1.3.05
9
Figure 7 - Price Sweep for Soyuz Purchase.
Reliability Sweep
Fig. 8 is a very interesting result. Here, the lower
the reliability, the better the business case. This
is because the project does better when it is
driven out of business early by the failure model.
Obviously, this should not be taken as
encouraging low launch vehicle reliability, but it
may indicate a proper time limit on this
particular project. This trade was done for a
constant $10M ticket price.
Figure 8 – Reliability Sweep.
Sub-orbital Reusable Rocket
The inclusion of this vehicle is designed to test
the feasibility of near-term sub-orbital Reusable
Launch Vehicles (RLV’s) at providing
entertainment class space transportation. When
compared to an orbital rocket of similar design,
the sub-orbital rocket is much smaller, with
lower up front and operating costs. It also
performs a less stressful mission profile than a
comparable orbital RLV.
Concept Assumptions
This vehicle is an X-Prize-class 10 passenger
sub-orbital reusable rocket.9 It has one rocket
engine for power and a wing-body configuration
using kerosene for fuel and liquid oxygen for
oxidizer. As it requires cargo aircraft
transportation to return to the launch site, this
amount is included in the launch site fee. It is
important to note that this is a zero-order
estimate and not a complete concept, but it
should be representative of this class of vehicle.
The engineering vehicle characteristics are given
in Table 1.
Table 1 – Sub-Orbital Vehicle Characteristics
Parameter Value
Gross Weight 265 klb.
Dry Weight 35 klb.
Vacuum Thrust 370 klb.
Sea Level Thrust 330 klb.
Mass Ratio 6.80
Figure 9 – Sub-Orbital Vehicle Three View.
Price Sweep
It is evident from Fig. 10 that there is an
optimum price at around $8M. This is not
surprising since there is a recurring cost
associated with this vehicle on the same order of
magnitude as this ticket price.
-1400
-1200
-1000
-800
-600
-400
-200
00 2 4 6 8 10
Reliability (# nines)
NPV ($M, 15% disc ount rate)
10th Precentile NPV
Mean NPV
90th Percentile NPV
-1000
0
4
5
6
7
8
9
10
Price ($M)
NPV ($M, 15% discount rate)
-2000
-3000
-4000
-5000
-6000
IAA-00-IAA.1.3.05
10
Figure 10 – Price Sweep for Sub-Orbital Rocket.
Reliability Sweep
Figure 11 – Reliability Sweep for Sub-Orbital
Rocket.
The reliability sweep at a constant ticket price of
$8M for this vehicle shows a more conventional
set of curves than that of the Soyuz purchase
plan. It appears that a vehicle of this type will
need 99.99% (four nines) reliability in order to
avoid economic penalties for failure. All three
confidence levels seem to follow the same trend.
Second Generation RLV Add-on Module
There is a chance that in the near future, there
will be a commercial RLV with the capability to
return payload from orbit. If the reliability of this
RLV is high enough, a low cost option for space
tourism might be to use this existing platform
with the addition of a passenger pod, or
SpaceCab. This concept represents minimal up-
front cost with low recurring cost for an orbital
vehicle.
Concept Assumptions
SpaceCab uses a 2nd Generation (RLV) to carry a
specially designed passenger cabin in its payload
bay, similar to the way the Space Lab module
rides in the payload bay of the Space Shuttle.
The defining characteristics for this module are
the number of passengers and total time on
internal power. The number of passengers is
determined by a gross mass constraint of 40 klb.,
the estimated payload capacity of a typical 2nd
Generation RLV concept. Based on these
weights, development costs are estimated at 912
M$ DDT&E and 208 M$ TFU. Because of this
additional financial risk, the discount rate is
20%.
Figure 12 – Three View of Example Space Cab.
Price Sweep
The pricing information for the SpaceCab in Fig.
13 concept seems to indicate the higher, the
better. From this graph, an optimum ticket price
of $10M is selected. This is partly due to its
positive NPV and partly due to it low NPV
variance.
Figure 13 – Price Sweep of RLV Add-on Module.
Length 65 ft
Diameter 15 ft
-600
-500
-400
-300
-200
-100
0
100
200
300
400
0 2 4 6 8 10 12
Reliability (# nines)
NPV ($M, 20% discount rate)
10th percentile
Mean
90th percentile
-2500
-2000
-1500
-1000
-500
0
500
1000
0 2 4 6 8 10 12
Price ($M)
NPV ($M, 20% discount rate)
10th percentile
Mean
90th Percentile
0
20
40
60
80
100
120
4 5 6 7 8 9 10 11
Price ($M)
NPV ($M, 20% discount rate)
10th percentile
Mean
90th percentile
IAA-00-IAA.1.3.05
11
Reliability Sweep
The curves for reliability in Fig. 14 show that the
concept is fairly insensitive to the possibility of
failure. This is most likely due to its low flight
rate and high ticket price. Only when the chance
of failure is greater than one percent does the
NPV begin to suffer.
Figure 14 - Reliability Sweep of RLV Add-On
Module
Third Generation Dedicated RLV
An advanced third generation RLV was tested to
determine how well a dedicated space tourism
vehicle designed to ferry passengers to and from
low earth orbit would fare economically. This
vehicle has a considerable non-recurring cost
with low recurring cost. It also has a high level
of customer appeal, which helps the market
demand.
Concept Assumptions
Here a modified third generation launch vehicle
(Fig. 15) is considered.10 It is an RBCC-engined
SSTO vehicle with horizontal takeoff and
landing capability. It is assumed to be the
transportation segment of an orbiting space hotel
project and therefore has more market appeal
than a simple orbital vehicle.
For the business analysis in LMNoP, an
owner/operator is assumed for the launch vehicle
and the passengers pay the transportation
segment of their journey independently from the
hotel stay. This somewhat isolates the business
plan for the shuttle from the business plan for the
hotel.
Figure 15 – Advanced RLV Three View
Price Sweep
To get an idea of far future business
opportunities, an advanced RLV concept was
analyzed with LMNoP across a range of prices.
Apparently, the low recurring cost estimate for
this vehicle was not enough to overcome the
high nonrecurring costs. This vehicle loses
money for all price ranges relative to a 25%
discount rate.
Figure 16– Price Sweep of Advanced RLV
Reliability Sweep
Figure 17 – Reliability Sweep of Advanced RLV
0
20
40
60
80
100
120
0 2 4 6 8 10
Reliability (# nines)
NPV ($M, 20% discount ra te)
10th percentile
Mean
90th percentile
(16,000.0000)
(14,000.0000)
(12,000.0000)
(10,000.0000)
(8,000.0000)
(6,000.0000)
(4,000.0000)
(2,000.0000)
-0.00 2.00 4.00 6.00 8.00 10.00 12.00
Price ($M)
NPV ($M, 25% disc ount rate)
10th percentile
Mean
90th percentile
(3,000.0000)
(2,500.0000)
(2,000.0000)
(1,500.0000)
(1,000.0000)
(500.0000)
-0 2 4 6 8 10 12
Reliability (# nines)
NPV ($M, 25% disc ount rate)
10th percentile
Mean
90th percentile
IAA-00-IAA.1.3.05
12
At the constant price of $8M, it does not appear
that the reliability required is any different from
any other vehicle in this price range. Fig. 17
shows there is again a significant penalty for
going below 99%, but reliability above that is
more than able to support the flight rate.
ECONOMIC PARAMETER SCREENING
ARRAY
Purpose
To determine the economic drivers for a
successful space tourism business, a screening
array was conducted on the inputs to LMNoP.
These include the vehicle performance and cost
characteristics as well as the business scheduling
information, such as the amount of time for
DDT&E and time to build the first vehicle. This
test yields valuable information regarding where
cost cutting efforts should be directed in
commercial RLV technology for space tourism.
Procedure
The screening array used for this test was a 32
run, 2 level fractional factorial design for 24
variables. This test yields unconfounded first
order effect information with a small number of
highly confounded second order effects. The
final effect test was run both with and without
the two level effects and showed little difference
in the magnitude and ordering of the driving
factors. This indicates that there is probably little
interaction between the input variables.
The primary ranking criterion is the 80%
confidence-level on NPV. This was chosen
because it is a conservative measure of the
profitability of the project being screened.
Variables
The inputs variables for the screening arrays and
a brief definition of each are described below:
• Engine TFU - The theoretical first unit
(TFU) cost of the first operational engine of
the vehicle program. This value is
irrespective of any learning curve effect.
• Engine Life - The number of total flights
before replacement of an engine on the
vehicle is necessary.
• Engines/ airframe (AF) - The number of
engines per airframe for the vehicle.
• Equity market access count - The number of
rounds (years) during the life of the program
when equity in the commercial entity is sold.
Financing is accomplished by selling
common stock or preferred stock to
investors.
• Capital on hand - The amount of capital
possessed by the company at the beginning
of the project. This value is irrespective of
the project being evaluated for investment.
• Tax Rate - The governmental tax rate on the
commercial entity’s net income.
• Interest Rate - The basic value of the interest
rate for long-term debt for the commercial
entity (cost of debt capital).
• Equity financing frequency - The number of
years from one round of equity financing to
the next (if multiple offerings are desired)
starting from the second round of equity
financing.
• Equity-offering amount - The amount of
equity in the commercial entity sold in each
round (year) of financing.
• Fixed SG&A expense - Balance sheet item,
which combines base salaries, commissions,
and travel expenses for executives and
salespeople, advertising costs, and payroll
expenses per year.
IAA-00-IAA.1.3.05
13
• Variable SG&A expense - Balance sheet
item, which combines incremental salaries,
commissions, and travel expenses for
executives and salespeople, advertising
costs, and payroll expenses per launch.
• Time for DDT&E - The number of years
required for the vehicle airframe / engine
design, development, testing, and evaluation
(DDT&E).
• Time from Production to IOC - The number
of years from start of initial rate vehicle
airframe and engine production to initial
operating capability (IOC).
• Time to depreciate fixed assets - The
number of years used to depreciate all fixed
assets in the program.
• Passengers per Launch - The passenger
capability of the vehicle.
• Reliability - The overall system reliability of
the vehicle (includes airframe and engine.)
• AF life - The number of total flights before
replacement of the airframe on the vehicle is
necessary.
• Turn around time (TAT) - The number of
elapsed days it takes for a vehicle returning
from a mission to be recycled in preparation
for the next launch.
• Time in flight (TIF) - The number of elapsed
days for a typical vehicle mission.
• AF DDT&E - The cost for design,
development, testing, and evaluation
(DDT&E) of the airframe of the vehicle.
• AF TFU - The theoretical first unit (TFU)
cost of the first operational airframe of the
vehicle program.
• Engine DDT&E - The cost for design,
development, testing, and evaluation
(DDT&E) of the engine of the vehicle.
• Add-on contribution per launch - The
additional revenue per launch obtained
through non-primary sources.
• Customer Appeal - Multiplier placed on
baseline market demand to account for
factors such as comfort, flight duration and
visibility
Vehicle Test Variable Ranges
For the test on the near term sub-orbital and third
generation orbital RLV’s, the variables described
in the variables section were used. All monetary
values are for fiscal year 2000 (FY2000.) Their
levels for these tests were as follows:
Table 2 – Settings for Sub_Orbital RLV
Screening Array
Variable Low High
Engine TFU $6M $10M
Engine Life 75 flts. 125 flts.
Engines per AF 1 2
Equity market offerings 2 4
Capital on hand $1.5B $2.5B
Tax rate 0% 37.5%
Interest rate 7.5% 12.5%
Equity financing offerings 2 4
Fixed SG&A expense $22.5M $37.5M
Variable SG&A expense $100K $1M
DDT&E duration 2 years 4 years
Time for production 1 year 2 years
Time to depreciate assets 3 years 7 years
Passenger Capacity 8 12
Vehicle Reliability 0.99 0.9999
Airframe life 375 flts. 625 flts.
Turnaround time 5 days 7 days
Time in flight 0.5 days 1 day
Airframe DDT&E $2.25B $3.75B
Airframe TFU $750M $1.25B
Amount at equity offering $375M $625M
Engine DDT&E $0M $0.1M
Advertising fee $0 $0.5M
Market Appeal Factor 0.25x 0.5x
IAA-00-IAA.1.3.05
14
Table 3 – Variable Settings for Advanced RLV
Screening Array
Variable Low High
Engine TFU $62M $104M
Engine Life 375 flts. 625 flts.
Engines per AF 4 6
Equity market offerings 2 4
Capital on hand $1.5B $2.5B
Tax rate 0% 37.5%
Interest rate 7.5% 12.5%
Equity financing offerings 2 4
Fixed SG&A expense $22.5M $37.5M
Variable SG&A expense $100K $1M
DDT&E duration 3 years 5 years
Time for production 1 year 2 years
Time to depreciate assets 3 years 7 years
Passenger Capacity 23 27
Vehicle Reliability 99.9% 99.9999
%
Airframe life 750 flts. 1250 flts.
Turnaround time 5 days 7 days
Time in flight 1 day 3 days
Airframe DDT&E $5.78B $9.63B
Airframe TFU $1.1B $1.8B
Amount at equity offering $375M $625M
Engine DDT&E $333M $368M
Advertising fee $0 $0.5M
Market Appeal Factor 2x 8x
Results
Sub-orbital Reusable Rocket Variable Effects
The results for the sub-orbital RLV effect
screening are interesting. As expected, the cost
and scheduling variables are quite important to
the response. However, the major player is the
government tax rate. This is likely due to the fact
that the bottom value of the experiment design
for this variable was zero percent. Zero tax rate
would reflect a potential tax-free policy for space
tourism enterprises to help the industry get
started. It is important to note that these
rankings depend a great deal on the area of the
design space being explored.
Figure 18– Pareto Plot for 80% Confidence
NPV for Sub-Orbital RLV
IAA-00-IAA.1.3.05
15
Looking subjectively at this Pareto plot, the
major variable players are:
• Tax rate
• Number of years for DDT&E
• Airframe TFU
• Airframe DDT&E
• Add-on revenue per launch
• Number of years from unit production to
IOC
• Engine TFU
Engine TFU must be considered because of its
interaction with engines per airframe. This
information will serve as a guideline when
conducting the space tourism economic goal
search.
Third Generation RLV Variable Effects
The advanced RLV has customer appeal as its
major factor. This translates to increased
importance of the market prediction model
variance for this concept. It should be noted that
the overall effect of Engines/AF is to change the
cost values for the engines. Therefore, the
importance of all these variables can be
considered linked.
Again looking subjectively at the Pareto plot, the
major drivers are:
• Customer appeal
• Engines per airframe
• Number of years for DDT&E
• Engine TFU
• Turn around time
• Number of years from unit production to
IOC
Most of the other effects are likely due to noise.
Figure 19 – Pareto Plot for 80% Confidence
NPV for Advanced RLV
PRIORITIZED GOALS FOR SELECTED
CONCEPTS
Procedure
For this part of the research, the variable inputs
of LMNoP are changed until a viable space
tourism project is attained (defined as 80%
confidence of positive NPV.) This is done for the
purpose of identifying an example of what cost
goals will result in a viable project. Of course, it
must be said the settings that result in a viable
vehicle are not unique.
This is done for two vehicle projects. The first is
the near term technology sub-orbital rocket from
IAA-00-IAA.1.3.05
16
the screening array. This viability search is based
on changing the variable values from their
baseline values. This is possible because of the
near feasibility of the screening array results.
The test for the far term vehicle is somewhat
different. Using contemporary estimates, the
economic parameters for this vehicle were
insufficient to yield a workable concept. This
means the results of the screening array are not
valid for this low price, high flight rate scenario.
Sub-Orbital Rocket
Problem Statement
In order to ensure a reasonable final set of design
variables, an error function (Eqn. 3) has been
introduced. This function includes a reasonable
range for each variable to make sure that each
term is weighted properly.
(3)
Using this, the problem statement for this part of
the research is to minimize the Error function
while maintaining a viable design. To be viable,
all of the input variable settings must be
physically possible and the 80% confidence level
of NPV must be positive.
The variable set for this problem can be inferred
from the results in Table 4.
Results
Several large changes from the initial baseline
values were required to attain a positive NPV for
80% of the cases. The largest adjustment was the
Capital on hand. Higher capital on hand tended
to lower the spread on NPV by reducing the
chances of having financing costs dominate the
LCC.
Table 4 – Variable Setting Results of Goal
Analysis for Sub-Orbital RLV.
Variable Baseline Final
Engine TFU $8M $6M
Capital on hand $2B $5B
Tax Rate 30% 0%
Interest Rate 10% 7.5%
DDT&E duration 3 years 3 years
Production duration 1 year 1 year
Reliability 99% 99.9%
Airframe DDT&E $3B $1B
Airframe TFU $1B $200M
Add-on Contribution $0 / flt. $1M / flt.
Customer Appeal Sub-coach 1st class
Fig. 20, the final distribution of NPV, shows a
large spread, but 80% of the distribution is
positive. This shows that if these cost goals can
be met, there is a high probability of a project
like this succeeding.
Figure 20 – Final Distribution of NPV for Sub-
Orbital Rocket.
Third Generation RLV
The baseline values for the third generation RLV
did not provide any chance for this concept to
become feasible. Therefore, an example using
the assumption of low ticket price as well as
airline-like operations and recurring cost was run
as an example goal for this market segment.
∑
−−
=
iablesall asonableasonable baselinesettingVariable
Error
var
2
max_Remin_Re _
$4,000 $2,250
$500 ($1250) ($3000)
NPV ($M, 25% discount rate)
IAA-00-IAA.1.3.05
17
Assumptions
To attempt to simulate the performance of a far-
future space tour airline, some rather optimistic
assumptions were made. These are documented
below in Table 5. All dollar values are for
FY2000.
Table 5 – Third Generation RLV Optimistic
Assumptions
Variable Setting
Airframe DDT&E $20B
Airframe TFU $100M
Engine DDT&E $3B
Engine TFU $20M
Recurring Cost $10,000 per flight
Engines per airframe 4
Reliability 99.999999%
Airframe & Engine Life 3,000 flights
Fixed SG&A expenses $15M per year
Variable SG&A expenses $10,000 per flight
Turn around time 0.1 days
Time in flight 0.5 days
Launch site fee $10,000 per flight
Customer Appeal 1st class w/ Orbital
Hotel
Capital on hand $10B
Ticket Price $15,000 per seat
Passenger Capacity 27
Tax Rate 30% per year
Inflation Rate 3% per year
Cost of failure $200M
Results
Fig. 21 shows that the assumptions above do
provide for the possibility of a viable vehicle
according to the requirements of this test.
However, the variance of the NPV is so large
that it is still uncertain whether this business will
be boom or bust.
Figure 21 – NPV Results for Optimistic
Assumptions
The area to the left of the line in Fig. 19 has
negative NPV while the area to the right has
positive. The integrated probability of positive
NPV is 60%. An advanced RLV just for space
tourism appears to be quite a gamble.
CONCLUSIONS
The conclusions of this research cover the areas
of feasibility and technology areas for future
concentration. These should be considered as
recommendations.
1. Space tourism as a concept could be
feasible. With maturation of certain
technologies, there might be a concept
capable of supporting a feasible space
tourism business.
2. Large leaps in cost metrics will be required
to make space tourism a reality for the
average person. This type of operation
requires truly airline-like operation,
something out of reach for current launch
vehicle approaches.
3. Design and construction cycle times are
important to the feasibility of the concepts
observed here. This means that advanced
design and construction planning techniques
are just as important as other technologies to
the success of space tourism.
4. Government policy is vital to the growth of
this industry. Incubation policies are
important to the near term industries, while
NPV ($M, 25% discount rate)
Probability
IAA-00-IAA.1.3.05
18
strict safety guidelines will be needed as
flight rates rise.
FUTURE WORK
Several items for potential future work have been
identified during the course of this work.
1. LMNoP Market Model – The market model
in LMNoP randomly selects a point from an
uncertainty distribution every year. This
point is unrelated to the point selected for
the previous year. It would be more realistic
to assume that there is a large uncertainty
the first year, with small dispersions in
subsequent years. This large randomness in
demand causes problems with purchasing
schedules, etc. that would likely not be as
extreme in a real business.
2. Computational Speed – The computational
cost of the LMNoP spreadsheet is
significant. It currently consumes about one
hour on a 500 MhZ Pentium III to complete
a full Monte Carlo simulation of one
vehicle. This is a hindrance to trade studies
or optimization. There is a possible future
effort to translate CABAM11 (Cost and
Business Analysis Module, the Space
System Design Lab cost model) into a
compiled code. Since LMNoP and CABAM
share a few components, it might be possible
to also compile LMNoP with minimal effort.
3. Vehicle Design – A more in-depth vehicle
design process may yield new insight into
lucrative areas of the design space.
ACKNOWLEDGEMENTS
The authors would like to acknowledge NASA
Langley Research Center for their support of this
project under grant number NAG-1-2280. They
would also like to acknowledge the help of Matt
Medlin and Brad St. Germain, graduate students
in SSDL who have provided analysis support.
REFERENCES
1. Penn, J. P. and Lindley, C. A.,
“Requirements and Approach for a Space
Tourism Launch System,” IAF paper IAA-
97-IAA.1.2.08, October 1997.
2. Penn, J. P., “Synergy Between Space Power
and Low-cost High Flight Rate Reusable
Launch,” IAF paper IAF-98-R.2.09, October
1998.
3. Rogers, T. F., “Space Tourism: The
Perspective in Japan and Some Implications
For the United States,” The Space
Transportation Association, August 1994.
4. Smitherman D. and McClure, W., “Space
Transportation and Destination Facilities
Working Group Proceedings,” STA/NASA,
March 1997.
5. Commercial Space Transportation Study
Final Report, Boeing, General Dynamics,
Lockheed, Martin Marietta, McDonnell
Douglas and Rockwell, pp. 198-229, April
1994.
6. Nagatomo, M. and Collins, P., “A Common
Cost Target of Space Transportation for
Space Tourism and Space Energy
Development,” AAS 97-460, pp. 617-630.
7. Isakowitz, Steven J., “International
Reference Guide to Space Launch Systems
Second Edition,” AIAA, pp. 160-171, 1991.
8. Wade, M., “Encyclopedia Astronautica,”
Web Site Address: http://www.friends-
partners.org/~mwade/spaceflt.htm, 2000.
9. “X-Prize Foundation // New Race to Space,”
Web Site Address: http://www.xprize.org,
2000.
10. Olds, J., Bradford, J., Charania, A.,
Ledsinger, L., McCormick, D. and
Sorensen, K., “Hyperion: An SSTO Vision
Vehicle Concept Utilizing Rocket-Based
Combined Cycle Engines,” AIAA paper
number 99-4944, November 1999.
11. Whitfield, J. and Olds, J., “Economic
Uncertainty of Weight and Market
Parameters for Advanced Space Launch
Vehicles," AIAA 98-5179, October 1998.