Summary of Design
Sheet Features
Design Sheet™ is a systems design
and analysis tool directly applicable to the early stages
of design where, depending on the industry, approximately
seventy to eighty percent of the life cycle cost of a product
is determined. This 70%-80% cost factor holds true for
complete engineering systems involving mechanical and electromechanical
subsystems, such as aircraft or spacecraft, as well as
their individual components. Consequently, it is extremely
beneficial to be able to balance the concerns of competing
performance objectives and costs during this crucial early
stage. Optimal choices have a disproportionately dramatic
impact on the downstream manufacturing and support costs
relative to the optimizations effected during the detailed
design and manufacturing stages.
Design Sheet
is aimed directly at the conceptual designer, who must build
and then assess models with thousands of variables - often
in a very short period of time. It is unique in several ways.
First, it facilitates the designer by providing a model building
environment that supports the development of the large models
needed: the user interface is designed to allow the designer
to enter the design equations and quickly assess their mathematical
completeness and solve them rapidly. Second, Design Sheet's
unique constraint management technology allows the designer
to manipulate the model equations with complete flexibility:
any variable can be fixed at a desired value and then the other
variables calculated from it. For problems like aircraft design,
this tool allows the designer to target a specific cost and
then explore the design space that is relevant to that cost
only. This is called designing with Cost as an Independent
Variable, or CAIV; Design Sheet is the only industrial level
design tool capable of this.
Because
Design Sheet provides a design environment for flexibly and
rapidly building almost arbitrary complex engineering models,
the designer is encouraged to explore a significantly wider
range of alternatives in a given period of time, as well as
simultaneously consider multiple performance objectives, cost,
manufacturability, and reliability in a manner that was heretofore
impossible.
Design
Sheet's Constraint Management Features
Design Sheet's unique constraint management
technology is key to its ability to support simultaneous,
rapid trade-off studies of multiple aspects of the performance
and cost of a complex system. Under continuing development
at Teledyne
Scientific Company, Design Sheet
offers the analyst the ability to exploit models of thousands
of performance
parameters, acquisition cost elements, life-cycle costs,
and other critical system attributes by providing an extremely
flexible environment to explore the trade-off space represented
by these models.
This graphic
summarizes Design Sheet's features. For its inputs, Design
Sheet takes models relating the design parameters in a given
domain. It uses constraint management technology to manage
the resulting system of principally non-linear algebraic equations.
Key to Design Sheet's success in practice is the ability of
these algorithms to scale to the extremely large sets of equations
characteristic of complex engineering systems such as aircraft,
satellites, or space vehicles.
Design Sheet
has a specially designed interface to manage the system of
equations and allows the user to easily define new trade studies
on the fly, producing charts, graphs, and tables with a few
mouse clicks.
A unique
feature of Design Sheet is that it allows the user to change at
runtime which variables are input to the model. Thus,
he can make cost an input, allowing, for example, trade-offs
among different performance objectives for a fixed cost or
set of costs.
Example:
Hand Held Transceiver Design
Flexible
Analysis Tool
Trade
of cost and weight varying carrier
frequency and distance for S/N = 10 dB
The above
example shows a simple model of signal to noise as a function
of power, distance, etc., along with empirical models of cost
and weight. The joint models are managed by Design Sheet without
the user having to write any integration code. When S/N is
input to the model, the user can perform a trade-off of cost
and weight for fixed performance objective of S/N=10 dB.
Aircraft
System Design
In the early
stages of aircraft design, a major activity is aircraft "sizing" in
which estimates are made of the gross take off weight needed
to achieve simplified mission profiles. These analyses, coupled
with estimates of other performance attributes and cost assessments,
typically involve more than 1000 equations and variables. Shown
below is a trade study produced by Design Sheet for an F-16
like fighter aircraft. The trade assesses the impact of different
manufacturing processes, wing area choices, and range requirements
on Unit Fly-Away cost.
Cost
as an Independent Variable: Aircraft Design Example
The above
graphic shows a different view of the same design space represented
by the models used for the previous chart. Here, however, cost
has been set as an independent variable. The analysis shows
the impact of wing loading (gross take off weight divided by
area available for lift), range requirements, and wing area
on the structural load limit for two different input costs.
Designers
Must Balance Many Concerns
Most weapons
systems design problems must simultaneously consider many performance
objectives along with cost. Cost itself is a many faceted entity.
Typically, one must also be concerned with factors such as
flexibility and responsiveness of the system to meet different
needs. The top chart outlines the parameters in a model for
a fleet of trans-atmospheric vehicles. The constraint management
technology within Design Sheet is ideal for utilizing these
sorts of models. As long as variable names have the same meaning
in the different sub-models, no extra effort is needed in integrating
all models. The analyst can just include the constraints in
the same network and Design Sheet takes care of integration.
A
Complete Multi-disciplinary Design Example
The following presents a fairly complete
example of simultaneous performance and cost modeling.
The system being investigated is a UAV-based surveillance
system. This was work jointly performed by RSC, Boeing,
Lockheed-Martin, and Raytheon as part of the DARPA RaDEO
program.
Scenario: Starting with UAV mission requirements, with life cycle
cost considerations, demonstrate selected flow down requirements
for the preliminary aerodynamic structure, the IR sensor
subsystem and gimbal assembly, and their interactions. On
the basis of these subsystem requirements, perform cooperative
evaluations of design alternatives. Trade studies will include
producibility, process plans and their simulations.
The UAV
reconnassaince mission is to take off from a given air base,
reach a specifed altitiude (to be determined during the trade
studies), fly a given range (another trade parameter), loiter
on station for a given time (trade variable), and use an on-board
IR sensor to detect enemy SCUD missile launchers. If found,
the system should radio back to a ground contral station which
has fire control to attack the SCUD missile launch site. The
cost and performance of the system needs to be explored over
a wide range of design options.
Performance
and Cost Models: This system has been modeled using parametric
relationships for the performance and cost of the air vehicle,
on-board sensors, optics, and gimbaling systems. Mission
performance includes the survivability, endurance, and range
of the air vehicle along with probability of detection, field
of view, and sweep range of the on-board sensors. The aerodynamics
model includes simple lift and drag relationships used to
estimate fuel and weight requirements to meet range, altitude,
and endurance requirements. Both development and production
costs of the system are considered as a function of the design
attributes. The entire model contains approximately 800 relations
and 1000 variables. An example trade study is shown below:
Publications
- Constrained Exploration of Trade Spaces, (2006) SMCIT 2006, Space Mission Challenges for Information Technology, Pasadena, CA 2006.
- Planning Sensing Actions for UAVs in Urban Environments (2005) Proceedings of SPIE, Volume 5986, Unmanned/Unattended Sensors and Sensor Networks II, Edward M. Carapezza, Editor, 59860J, Oct. 26, 2005.
- Managing Function Constraints in Design Sheet (1998). Design Theory and Methodology Conference, September 1998, Atlanta, GA.
- Design Sheet: A System for Exploring Design Space, Application to Automotive Drive Train Life Analysis (1996) in J. S. Gero and F. Sudweeks (eds), Artificial Intelligence in Design ’96, Kluwer Academic Publishers, Netherlands, 1996, pp. 347-366.
- Constraint Management Methodology for Conceptual Design Tradeoff Studies (1996). Design Theory and Methodology Conference, August 1996, Irvine, CA
- Facilitating Infrared Seeker Performance Trade Studies Using Design Sheet (1995). Rockwell Palo Alto Laboratory technical report for Advanced Research Projects Agency of the Department of Defense through Wright Patterson Air Force Base under contract F33615-94-C-4426.
- Design Sheet: An Environment for Facilitating Flexible Trade Studies During Conceptual Design (1992). Aerospace Design Conference, February, Irvine, California
|
