ASRR Conference:
Engineered Resilient Systems Panel
Ensuring Engineered Resilient
Total-Ship Systems:
Using More Physics-Based
Design Tools in Early Concept Design
Bob Keane, Ship Design USA, Inc.
5 October 2011
Opinions are those of the author and not those of any Government agency or program.
Inseparable Resilience Issues
Dr. Robert Neches, OASD(R&E), 31 March 2011
 Rapidly Adapting to User Needs
– Warfighter / Engineer Information Exchange
 Accelerated Needs Exploration: Conceptual
Design
– Co-Evolution of ConOps & Systems for Multiple
Alternative Futures
 Accelerated Design-to-Build: Capability
Engineering
 Advances must be embedded in tools and
environments accessible to Government and
industry – Systems 2020
2
Nature of the Product-A Warship
•
•
•
•
•
•
•
•
A “system of systems”: Operate in battle force with older and newer ships
Very low quantities, high unit cost, long lives
Extremely complex product: millions of piece parts
No prototypes, first ship(s) must be fully operational
Government develops combat/weapons/communications systems
Government ultimately assumes responsibility for meeting requirements
Lack of commercial shipbuilding industrial base to build upon
Intense Congressional oversight
5/3-6/05
Summer Naval Surface Ship Design Program
3
USS MIDWAY 50-Year Global Crisis Response
Italy
Elections
1947
Korean Tree
Incident
1976
Iranian
Hostage
Crisis
1979
Afghanistan
Invasion 1979
Fiery Vigil
1991
Laos
1961
Lebanon
1950
5/3-6/05
Desert Shield /
Desert Storm
1990–91
QuemoyMatsu
1958
Earnest Will
1987–1988
Philippine
Coup
1989
Vietnam
1965–1975
Summer Naval Surface Ship Design Program
4
Collaborative Innovation in Fleet Design,
1922-1941
Naval War
Naval War
OpNav
OpNav
College
College
The General
The General
Board
Board
The Fleet
The Bureaus
The Bureaus
Agents of Innovation: The General Board and the Design of the Fleet That Defeated the
Japanese Navy, John T. Kuehn, Naval Institute Press, 2008
MIT'S VON HIPPEL: INNOVATION, LEAD USER
ANALYSIS, CUTTING CONCEPT TIME & COST
 “…sometimes it's a waste of time and
resources to try to understand customer
needs…
 sometimes it's less expensive and more
efficient to
– let your customers define the needs,
– limit yourself to offering solutions, and
– let the customers design based on that.”
6
Experience and the Right Tools
Make A BIG Difference
Lack of
Experience
Highly
Experienced
Smart, PhysicsBased Tools
Tools
Smart, PhysicsBased Tools
 1980s Reagan Buildup:
Average Cost Growth for
Lead Ship was 10%**
– NAVSEA Highly
Experienced Ship
Design Workforce
– User-Driven Design
 1990/2000s Acquisition
Rule-Based Tools
Rule-Based Tools
Lack of
Experience
Highly
Experienced
Experience
Reform: Average Cost
Growth for Lead Ship was
50%**
– Inexperienced Industry
Design Teams
– Mfr.-Driven Design
** SEA 05C, June 2008
Seaway Loads for Design of Surface
Combatants: Rule-Based Design
 Structural Design of FFG 7, CG 47, DDG 51 Classes
– Interested more in extreme loading conditions
than actual loads which contribute to fatigue
– Worked with simplified loading envelopes
– Deterministic analysis resulted in scantlings for
maximum load expected
– Highly random wave-induced loads were set of
simplified hydrostatic loads under extreme seas
No Physics-Based Computations nor Seakeeping
Model Tests to Determine Actual Seaway Loads
8
Lack of Physics-Based Design Tools:
Increased Ownership Costs
 FFG 7 Class
– Hull girder doubler plates & ballast added
– Extensive deckhouse fatigue cracking
 CG 52 Class (with VLS)
– Serious hull cracking and buckling problem
– Extensive superstructure fatigue cracking
 DDG 51 Class
– Bow structure buckling and cracking issue
 Operational loads exceeded rule-based design loads
$100M’s in repairs for sustaining service-lives
9
“Outside-In Design” – Start with Hull
Form, Then Cram Everything into Hull
 Hull is sized and shaped in early design based on:
– unreliable weight and area/volume estimates
– invalid assumption volume is “arrangeable”
– fallacy that limiting hull size limits ship costs
Need architectural tools to optimize arrangements
10
Ships Possessing Greater
Density Increase Production Cost
1200
Normalized First Ship Production (Hours / LT)
DDG
1000
Combatant Designs
FFG
800
CG
Future DDG
LPD
600
LHA
LHD
Auxiliary / Amphibs Designs
Future L-Ship
LSD
DISTRIBUTED
400
LOG-RO/RO
RO/RO
200
LMSR
AOE
AVIATION
Hybrid Designs
FLO/FLO 3
FLO/FLO 1 FLO/FLO 2
Legacy Hulls (Return Data)
MPF(F) Notional Designs (Estimated Data)
Commercial Designs
0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
Outfit Density (lbs / cu ft)
Ship Production hours increase with density and fall into
predictable groupings.
11
WHERE WE NEED TO GO
 According to an old
proverb, if we do
not change our
direction, we might
end up where we
are headed.
12
Systems Engineering –
Ensure “Elegant” Designs*
 Effective – it does what it is supposed to do
 Efficient – to produce, operate, maintain
 Robust - insensitive to variations in operations
 Minimal Unintended Consequences –
few band aids required to fix it in-service
*M. Griffin, Former NASA Director, Dean’s Seminar, SIT,
“How do we fix System Engineering?”, 13 Dec 2010
13
Toward Robust Systems Engineering:
CREATE-SHIPS Project
 Computational Research & Engineering for Acquisition
Tools & Environments (CREATE):
– Replace empirical design with validated physicsbased computational design
– Detect and fix design flaws early in design process
– Develop optimized designs for new concepts
– Begin system integration earlier in acquisition
process
– Increase acquisition program flexibility and agility
to respond to rapidly changing requirements
DoD High Performance Computing Modernization Program
14
Integrated Hydrodynamics Design
Environment - IHDE
RESISTANCE
POWERING MANEUVERING
LOADS
SEAKEEPING
SHAPE OPTIMIZATION
DESIGN STUDIES
LOCAL
HPC
HPCMP
AUTOMATED VALIDATION CASES
SSF
TSD
AEGIR
DAS BOOT
Problem
Set Up
SMP
VERES
Conditions
FREDYN
TEMPEST
Geometry
NavyFOAM
LAMP
Automated
Gridding
DRIVER/GUI
ASSET/LEAPS
15
Rapid Ship Design Environment
(RSDE) Architecture
Design
Data Sets
Rapid Ship Design Environment
Multi-Discipline Optimization
ASSET
Synthesis
SHCP-L Tkt
Hullform
Gen/Trans
Hulltran Tkt
Structure
Geometry
Hullform
Geometry
Stability
Behaviors
Seakeeping
Behaviors
NURBS
Math
4/13/20154/13/2015
Radial Basis
Math
GUI
Execution Engine
Behavior
ObjectZ
Design Space Exploration
SHCP-L
Stability
LEAPS API and Toolkits
ISA
Arrangement
Hullgen Tkt
IHDE
Hydro
BehaviorMan Tkt
LEAPS Database
Arrangement
Geometry
Subdivision
Geometry
Resistance
Behaviors
Structures
Behaviors
Flooding
Behaviors
GEML Kernel
Kriging
Math
Neural Net
Math
16
16
Modeling Warships in the
Ocean Battle Space: Way Ahead
 Need to start co-evolution of ConOps & systems for
multiple alternative futures long before Milestone A
– Continuing process not tied to specific acquisitions
 Need engineers to have unfettered access to warfighters
– Accelerated exploration during concept formulation
– Collaborative process exploring new/radical innovations
 Need more physics-based tools for concept design with
– Timely pre-processing of system geometry
– Timely post-processing of results into decision aids
 Need to reconstitute DoD S&T infrastructure
– DoD Labs resume their advanced development role
17
Center of Innovation in Ship Design is:
The hub of a national collaborative enterprise combining the
best ideas and experience of government, industry, and
academia in ship design
CISD Mission: Ensure the Future Capability (People, Tools
and Knowledge) of the Nation to Develop Innovative Ship
Designs to Effectively Meet Defense Needs
Focus Areas: Navy of the Future
•
•
•
•
Develop Future Ship Designers
Knowledge Base / Design Tools & Processes
Future Ship & Ship Design Technology Needs
Develop Innovative Ship Design Concepts
Chartered by ONR, NAVSEA 05 & NSWC
CISD Overview - March 2010
APPROVED FOR PUBLIC RELEASE
18