Instrument First, Spacecraft Second: A New Mission Development Paradigm Bob Bitten, Eric Mahr The Aerospace Corporation Claude Freaner NASA Headquarters, Science Mission Directorate 2011 NASA Program Management Challenge Long Beach, California 9-10 February 2011 sed with permission
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Instrument First, Spacecraft Second:A New Mission Development Paradigm
Bob Bitten, Eric Mahr The Aerospace Corporation
Claude FreanerNASA Headquarters, Science Mission Directorate
2011 NASA Program Management ChallengeLong Beach, California9-10 February 2011
Used with permission
2
Executive Summary
• Instrument development difficulties have been shown to be a significant contributor to overall mission cost and schedule growth
• An approach that starts instrument development prior to mission development, entitled “Instrument First, Spacecraft Second” (IFSS), could potentially lead to a reduction in cost growth
• An assessment of the IFSS approach was conducted looking at historical instrument development times to assess schedule variability at the mission level and its effect on a portfolio of missions
• Applying IFSS approach to the Tier 2 and Tier 3 Earth Science Decadal Survey (ESDS) missions has the potential to save NASA several billion dollars while providing additional benefits including:
– Launching full set of ESDS missions sooner– Increasing number of missions launched by a given date– Decreasing number of Threshold Breach instances
3
Agenda
• Background
• Approach Overview
• Individual Mission Simulation Results
• Mission Portfolio Simulation Results
• Considerations
• Summary
4
Background
• Observations– >60% of missions experience developmental issues with the instrument– These issues lead to increased cost for other mission elements due to
“Marching Army” cost– Recent missions such as ICESat, OCO & Cloudsat all had instrument
development issues• Results show instrument cost growth influences total mission cost
growth at 2:1 factor– Missions in which the instruments were almost fully developed, such as
QuikTOMS and QuikSCAT, were developed at minimal cost and on short development schedules while experiencing limited cost growth
• Hypothesis– Developing instruments first and bringing them to an acceptable level of
maturity prior to procuring the spacecraft and initiating ground system development could provide an overall cost reduction or minimize cost growth
5
Instrument Development Problems Account for Largest Contributor to Cost & Schedule Growth*
• Cost & Schedule growth data from 40 recently developed missions was investigated
• 63% of missions experienced instrument problems leading to project Cost and Schedule growth
• Missions with Instrument technical problems experience a much larger percentage of Cost & Schedule growth than missions with Spacecraft issues only
S/C Only22.2%
Other14.8%
Inst. Only33.3%
Both Inst & S/C29.6%
Distribution of Internal Cost & Schedule Growth
* As taken from “Using Historical NASA Cost and Schedule Growth to Set Future Program and Project Reserve Guidelines”, Bitten R., Emmons D., Freaner C., IEEE Aerospace Conference, Big Sky, Montana, 3-10 March 2007
24.1%
17.4%
9.3% 8.0%
18.7%
4.7%
34.6%
51.3%
0%
10%
20%
30%
40%
50%
60%
Cost Schedule
Pe
rce
nt
Gro
wth
Inst only
S/C only
Both
Other
Cost & Schedule Growth Due to Technical Issues
Historical NASA Data Indicates Payload Mass and Cost Growth Significantly Greater than Spacecraft Mass & Cost Growth
60%
101%
33%
44%
0%
20%
40%
60%
80%
100%
120%
Mass Cost
Ave
rage
Per
cent
Gro
wth
from
Pha
se B
Sta
rt
Payload
Spacecraft
6
1 1
Note: 1) As measured from Current Best Estimate, not including reserves
Data Indicated Payload Resource has Greater Uncertainty than Spacecraft
* As taken from “Inherent Optimism In Early Conceptual Designs and Its Effect On Cost and Schedule Growth: An Update”, Freaner C., Bitten R., Emmons D., 2010 NASA PM Challenge, Houston, Texas, 9-10 February 2010
Historical Instrument Schedule Growth*
< 0%
0 to 15%
15% to 30%
30% to 60%
> 60%
7
12%
30%
14%
30%
14%
Distribution ofInstrument Schedule Growth
Average Instrument Development Schedule Growth = 33% (10 months)
* Based on historical data of 64 instruments with non-restricted launch window
0
10
20
30
40
50
60
70
80
90
100
0 20 40 60 80 100
Planned Delivery Duration
Ac
tua
l De
live
ry D
ura
tio
n
Planned vs. ActualInstrument Development Duration
Cost* & Schedule Growth Examples
1.61.7
2.2
0.0
0.5
1.0
1.5
2.0
2.5
OCO CloudSat ICESat
Mis
sio
n t
o In
stru
men
t C
ost
Gro
wth
Rat
io
8
1.31.5
2.2
0
0.5
1
1.5
2
2.5
OCO CloudSat ICESat
Total Mission to InstrumentCost Growth Ratio
Instrument Schedule GrowthPlanned to Actual Ratio
* Note: Although it is understood that other factors contributed to the cost growth of these missions, it is believed that the instrument delivery delays were the primary contributor
Ratio of Mission Cost Growth to Instrument Cost Growth is on the order of 2:1
9
Case History: QuikSCAT
• On November 19, 1997, NASA awarded the first rapid spacecraft delivery order to Ball Aerospace & Technologies Corp., Boulder, CO for the delivery of QuikSCAT spacecraft
– The satellite was the first obtained under NASA's Indefinite Delivery/Indefinite Quantity program for Rapid Spacecraft Delivery Office (RSDO) for rapid delivery of satellite core systems
• QuikSCAT, NASA’s ocean-observing satellite mission, was rapidly developed to fill in the data gap between NSCAT on ADEOS-I and SeaWinds on ADEOS-II
– A scatterometer nearly identical to SeaWinds was quickly assembled from NSCAT spare parts
• QuikSCAT was launched on June 19, 1999 on a Delta II
Demonstrates that a 2-year procurement and S/C development, when instruments are complete, is feasible
10
Case History: QuikTOMS
• In July 1999, NASA selected Orbital Sciences Corporation (Orbital) to build, launch and operate the Quick Total Ozone Mapping Spectrometer (QuikTOMS)
– The fifth TOMS instrument flight model 5 (TOMS FM-5) was complete– FM-5 was originally scheduled to fly as a cooperative mission with Russia in late 2000 but was
delayed due to Russian funding issues, so it was decided to launch in August 2000 as a US free-flyer
– Named QuikTOMS since the effort entailed the construction and launch of a spacecraft in less than two years as compared to traditional missions which take from three to five years
• QuikTOMS was procured by NASA’s Goddard Space Flight Center’s (GSFC) Rapid Spacecraft Development Office (RSDO) and was managed by the GSFC QuikTOMS Project Office
– QuikTOMS, with the already built TOMS FM-5, was co-manifested as a secondary payload with Orbview 4
– Orbview 4, the primary payload, experienced integration and test difficulties, which caused a launch delay
• QuikTOMS was launched on September 21, 2001 on a Taurus
From FY03 Budget Document, pg. SAT 3-86, dated Feb-02
Demonstrates that a 2-year procurement and S/C development, when instruments are complete, is feasible
11
Agenda
• Background
• Approach Overview
• Individual Mission Simulation Results
• Mission Portfolio Simulation Results
• Considerations
• Summary
12
IFSS Development Approach Overview
Historical Development Approach
Instrument First, Spacecraft Second (IFSS) Approach
Spacecraft Development
Instrument Development
System I&T
Delay
System I&T
Marching Army
Plan Actual
Spacecraft Development
Instrument Development
System I&T
Delay
IFSS Offset
Comparison of Element Delivery Times – HyspIRI-like Mission
Schedule SimulationIFSS ResultsSand Chart ToolMeasures of
Effectiveness
• Cost to implement Tier 2 & 3 missions• Time to launch all Tier 2 & 3 missions• Number of missions launched by 2024• Percent of Threshold Breach Reports
Cost in FY10$M IndependentCategory EstimateMission PM/SE/MA 40.5$ Payload PM/SE/MA 7.3$ VSWIR 91.0$ TIR 54.7$ Spacecraft 94.4$ MOS/GDS Development 29.8$ Development Reserves 103.0$
Total Development Cost 420.7$ Phase E 24.2$ Phase E Reserve 4.0$ E/PO 1.9$ Launch System 130.0$
Total Mission Cost 580.7$
0.0%
10.0%
20.0%
30.0%
40.0%
50.0%
60.0%
70.0%
80.0%
90.0%
100.0%
300 400 500 600 700 800 900
Cum
ulati
ve P
roba
bilit
y
Estimated Cost (FY10$M)
Distribution
Sum of Modes
70th Percentile
HyspIRI-like Development Cost Risk Analysis Results –Case 1A, 1B & 2B (IFSS with 18 Month Offset) FY10$M
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
$200 $300 $400 $500 $600 $700 $800 $900
Estimated Development Cost (FY10$M)
Cu
mu
lati
ve
Pro
ba
bili
ty
Probability of Instrument Delaying Project• 96.7% for Case 1B no IFSS offset (9.8 month average delay)• 5.9% for Case 2B with 18 month offset
14
IFSS Assessment Overview
• Start with instrument resources– If no detailed instrument data can be found, then surrogates are used
• Size spacecraft for orbit conditions and instrument resource requirements
• Estimate the cost of the system
• Lay out baseline plan
• Phase cost over funding profile
• Identify analogous instrument development times to use in simulation
• Run the individual mission simulation
• Fold the mission simulation results into the mission portfolio simulation
Example Mission Data - HyspIRI Mission Overview
15
* Note: As taken from page 3 of HyspIRI presentation at Earth Science Decadal Survey Symposium, Feb 2009 , http://decadal.gsfc.nasa.gov/Symposium-2-11-09.html
Data Completeness Assessment
• Given the desire to have representative (i.e., “-like”) missions, surrogate instruments used when actual data was not available
16
Altitude Inclination Design Life
Mass Power Data Rate Duty Cycle Type
Tier 2HySPIRI X X X X X X X XASCENDS X X X P P P P XSWOT X X X P P XGEO-CAPE X X X P XACE X X X XTier 3LIST X X X X X P X XPATH X X X X X X X XGRACE-II X X X X
SCLP X X X X
GACM X X X X3D-Winds X X X X X X X X
Instrument ParametersMission ParametersMission
X = YesP = PartialBlank = No
Mission Concept Sizing
• Using mission and instrument parameters, representative Tier 2 and Tier 3 designs were developed
• Designs were developed using a Concurrent Engineering Methodology (CEM) model
• CEM model is a spreadsheet spacecraft conceptual design and analysis tool
– Sizing relationships generated using historical trend data
• Include physics, rules-of-thumb, parametric relationships, and educated guesswork
– Will not give an exact result, but provides representative designs “in the ballpark”
17
Comparison of Tier 2 & 3 Mission Public Costs vs. Estimate
18
MissionPublic Cost*
(FY10$M)
Aerospace Estimate(FY10$M)
Difference
Tier 2
HySPIRI-like 433$ 451$ 4.2%
ASCENDS-like 455$ 510$ 12.1%
SWOT-like 652$ 808$ 24.0%
GEO-CAPE-like 1,238$ 677$ -45.3%
ACE-like 1,632$ 1,285$ -21.2%
Tier 2 Total 4,409$ 3,731$ -15.4%
Tier 3
LIST-like 523$ 683$ 30.7%
PATH-like 459$ 387$ -15.7%
GRACE-II-like 454$ 280$ -38.3%
SCLP-like 449$ 552$ 22.9%
GACM-like 988$ 830$ -16.0%
3D-Winds-like 760$ 856$ 12.6%
Tier 3 Total 3,632$ 3,587$ -1.2%
Total 8,042$ 7,319$ -9.0%
Note: Costs do not include launch vehicle cost* Taken from NASA Day 2 - Earth Science and the Decadal Survey Program, Slide 20 February 2009 and inflated to FY10$,http://decadal.gsfc.nasa.gov/Symposium-2-11-09.html
Tier 2 Missions
Tier 3 Missions
Total
Results indicate that estimates are representative
19
Agenda
• Background
• Approach Overview
• Individual Mission Simulation Results
• Mission Portfolio Simulation Results
• Considerations
• Summary
Simple Schedule Analysis Simulation Framework
20
Instrument Development Delays Can Lead to Overall Schedule Delay
Spacecraft Development
Spacecraft Integration & Test
System Integration
Env. Test
Pad Ops.
Launch
Instrument Development
Instrument Integration & Test
SIR TRR
Typical DeliveryWith
Instrument Delay
Simulation of IFSS Approach
• If Instrument Dev + I&T to S/C > S/C Dev + System Integration Time– Add project marching army cost until instrument is complete
• If S/C Dev + System Integration Time > Instrument Dev + I&T to S/C– Add instrument marching army cost after instrument is developed
21
System ATP to TRR
Instrument ATP to Integration
}Cost due to Instrument Delay
System ATP to TRR
Instrument ATP to Integration
}IFSS Offset
}
Cost of Early Instrument Delivery
Instrument Delays Much More Costly than Early Instrument Delivery due to Marching Army
Example of Spacecraft & Instrument Timelines
• Basis of Triangular Schedule Distribution:– Low: Baseline Plan– Mode: Baseline Plan (S/C) and Average of Historical Analogies – High: Maximum of Historical Analogies
Spacecraft InstrumentDistribution ATP-TRR ATP-DelLow 45.0 44.6 Most likely 45.0 53.4 High 57.0 66.3 Mean 49.0 54.8
49 54.75340 45 50 55 60 65 70
Schedule Distributions (months)
Spacecraf t ATP-TRR
Instrument ATP-Delivery
}
Differences in means will lead toS/C waiting for instrument delivery
}22
Comparison of Element Delivery Times – HyspIRI-like Mission
45
40
44
10
13
4
12
16
8
20 30 40 50 60 70
TIR
VSWIR
Spacecraft
Months to Delivery
Minimum
Mean
Maximum
23
TIR instrument delivery time exceeds Spacecraft delivery time
Current Plan
Mission Simulation Overview
• To test the potential impact of implementing an IFSS approach, an analysis was conducted using historical instrument development durations to simulate the development of a mission
• A simulation was developed in which a Monte Carlo draw is made for both the spacecraft development duration and instrument development duration(s) to determine if the spacecraft will be ready for system testing prior to the instruments’ availability for integration to the spacecraft
– Simulation provides a statistical distribution of potential outcomes allowing for an assessment of the benefit or penalty of different IFSS offsets
• Two primary cases were studied – – Case 1: Baseline without any IFSS “offset”– Case 2: IFSS with an IFSS “offset”
24
Summary of Cases
• Case 1A – Plan without IFSS– Normal NASA mission development which has concurrent instrument,
spacecraft, and ground system development, with no unanticipated problems
• Case 1B – “Actual” without IFSS using Historical Data– Baseline with historically representative technical difficulties
• Case 2A – Plan with IFSS– “Instrument first" - development of instruments through successful CDR
and environmental test of an engineering or protoflight model prior to initiation of spacecraft and ground system development, with no unanticipated problems
• Case 2B – “Actual” with IFSS using Historical Data– “Instrument first" with historically representative technical difficulties
25
HyspIRI-like Development Cost Risk Analysis Results – Case 1A & 1B FY10$M
26
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
$200 $300 $400 $500 $600 $700 $800 $900
Estimated Development Cost (FY10$M)
Cu
mu
lati
ve
Pro
ba
bili
ty
Case 1BEstimate with
Instrumentdifficulties
$547M
Case 1AEstimate without instrument issues
$459M
HyspIRI-Like Development Cost Risk Analysis Results – Case 1A, 1B & 2B (IFSS with 18 Month Offset) FY10$M
27
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
$200 $300 $400 $500 $600 $700 $800 $900
Estimated Development Cost (FY10$M)
Cu
mu
lati
ve
Pro
ba
bili
ty
Case 1BEstimate with
Instrumentdifficulties
$547M
Case 1AEstimate without instrument issues
$459M
Case 2BEstimate with
Instrumentdifficulties
$466M
Probability of Instrument Delaying Project• 96.7% for Case 1B no IFSS offset (9.8 month average delay)• 5.9% for Case 2B with 18 month offset
Domino Effect is much greater leading to more inefficiencies & less funding available for future missions
Comparison of Mission Portfolio Results
$9.1
$11.1
$0.0
$2.0
$4.0
$6.0
$8.0
$10.0
$12.0
w/IFSS w/o IFSS
To
tal C
os
t F
Y1
0$
B
37
2024.1
2025
2023
2024
2025
2026
w/IFSS w/o IFSS
10.18.9
0
2
4
6
8
10
12
w/IFSS w/o IFSS
11.8%
64.2%
0%
10%
20%
30%
40%
50%
60%
70%
w/IFSS w/o IFSS
Cost to Implement ESDS Missions Time to Launch ESDS Missions
Number of Missions Launched by 2024 Percent Threshold Breach Reports
IFSS Provides Better Results for Each Metric Assessed
38
Agenda
• Background
• Approach Overview
• Individual Mission Simulation Results
• Mission Portfolio Simulation Results
• Considerations
• Summary
IFSS Considerations
• Typical IFSS “Offset” for instrument development is two years– Provides instruments with a two year head start prior to a three to four year mission
development phase
• For most instrument development efforts, this is after CDR but prior to full instrument integration
– At this point, most instrument problems should be identified– Time remains to recover prior to delivery to spacecraft for system environmental
test
• Assumes that mission systems engineers and spacecraft vendors are involved at low level of effort to ensure mission requirements and spacecraft accommodations are considered
• IFSS approach may not be suitable for all mission types– May not apply when spacecraft is integral to instrument
39
Rapid III Procurement* Can Provide Reliable Spacecraft with Known Performance within 20 to 36 Months
Typical 2-3 year procurement for spacecraft plus additional year for testing plus 2 year IFSS offset equates to 5 to 6 year total mission development time
* Note: As taken from Rapid III Spacecraft Summary, posted April 1, 2010, http://rsdo.gsfc.nasa.gov/Rapid-III.html
41
Agenda
• Background
• Approach Overview
• Individual Mission Simulation Results
• Mission Portfolio Simulation Results
• Considerations
• Summary
42
Summary
• Historically, Instrument development difficulties have been shown to be a significant contributor to overall mission cost and schedule growth
• An approach that starts instrument development prior to mission development, entitled “Instrument First, Spacecraft Second” (IFSS), could potentially lead to a reduction in cost growth
• Applying IFSS approach to the Tier 2 and Tier 3 Earth Science Decadal Survey (ESDS) missions has the potential to save NASA on the order of $2B while providing additional benefits including:
– Launching full set of ESDS missions a year sooner– Providing for an extra mission launched by 2024– Decreasing the number of Threshold Breach instances from 64% to 12%
• IFSS approach is enabled/enhanced given Rapid III Rapid Spacecraft Development Office (RSDO) bus procurement approach
– Availability of wide range of busses provides quick acquisition of required capability
Questions?
• Bob Bitten, NASA Advanced Projects, The Aerospace Corporation– [email protected]
• Eric Mahr, Space Architecture Department, The Aerospace Corporation– [email protected]
• Claude Freaner, Science Mission Directorate, NASA Headquarters– [email protected]