Doe in Racing

Design of Experiments in Racing
Design of Experiments in Racing

Carlo
Carlo
Rendon
Rendon
Statistical Methodology
Use a statistical approach to develop
meaningful conclusions from the data,
particularly when the data is subject to
external noise or experimental error
Statistical methods allow the engineer
to make recommendations and draw
conclusions to a reasonable amount of
certainty
Block Diagram
I n p u t s O u t p u t
. . .
x x x
1 2 p
2 1
z z
q
z
C o n t r o l l a b l e F a c t o r s
U n c o n t r o l l a b l e F a c t o r s
y
Advantages of DOE
Determine which variables are most
influential on the response, y
Determine where to set the influential xs so
that y is almost always near the desired
nominal value
that variability in y is small
that the effects of the uncontrollable variables
z1, z2, , zq are minimized
Benefits to FSAE
Given the large amounts of variation in a number of
operating parameters, it becomes difficult to weigh
the benefits of adjusting a single variable
When used appropriately, DOE can generate specific
experiments to assess the outcome response
produced by varying a single parameter within a
system
Efficient process that allows the experimenter to limit
the amount of test trials necessary to produce the
required data, thus reducing overall testing time
Benefits Continued
Can be used to generate the necessary data in a
minimal amount of testing time, or to improve on an
existing experimental process, saving time and
money
Used in racing to determine which parameters
produce the greatest effect on the system response
Traditional testing methods combined with statistical
design of experiments can be a great way to balance
cost, schedule, and performance
Procedure
Design the experiment choose factors and
levels and factorial design
Enter the response data complete
experiments in random run order
Analyze the results identify the factors that
produce the greatest effect on system
response and view statistics in detail
Statistical analysis and data plots view
graphs and determine desired factor levels
Factors and Levels
Excellent Poor Categorical Track Conditions F
Aggressive Conservative Categorical Driving Style E
Race Only Stock Street Categorical Tires D
Air Shifter 4.06 Low Categorical Shifter C
Race
Engine
Track Categorical Engine B
3500 lbs 2500 lbs Numeric Vehicle Weight A
High (+) Low (-) Type Name
Experiments vs. Factors
Experimentation
Trials carried out in Drag 2000 for two-
response system (1/4 mile time and top
speed)
Actual experimental trials should be
performed in the random run order provided
by the software
Simulating the experiments using software is
different from performing the actual trials due
to noise
Half Normal Probability Plot
mile time analysis
Interaction Graph
mile time analysis
Interaction Graph
mile time analysis
Half Normal Probability Plot
mile top speed analysis
Interaction Graph
Interaction Graph
Design-Ease
Point Prediction
The software analyzes trends in behavior of the data
and predicts the outcome for various parameter
changes
Results
Excellent Either Race
Only
Either Race
Engine
2500
lbs
Top Speed
(167.99 mph)
Excellent Aggressive Race
Only
4.06
Low
Race
Engine
2500
lbs
Time
(8.90 s)
Track
Conditions
Driving
Style
Tires Shifter Engine Weight
Factor levels for shifter and driving style had no effect on
the top speed - suggests that in designing a racecar for
speed, these factors are of less importance
The factors do provide a significant difference when
designing for the best time
Conclusion
Statistical design of experiments provides a method
of testing a large number of parameters at one time
to show which have the greatest effect on system
response, singling out those that produce a particular
desired response
Statistical methods cannot prove that a factor or
factors have a particular effect, they can only provide
guidelines as to the reliability and validity of results
Statistics allow one to measure the likely error in a
conclusion or to attach a level of confidence to a
statement
Conclusion
Tires, vehicle weight, and engine had the greatest
effect on the experimental outcome, although the
order of importance differed for each of the two
responses
Interesting observations were made regarding factor
interactions and design considerations for mile
time versus top speed
References
Be Bold. Math Options, Inc. 12 Dec. 2003. <http://www.mathoptions.com>.
Design-Ease
Software Users Guide. Version 6. Stat-Ease
, Inc., 2000.
Design of Experiments: Case Studies & Articles. Stat-Ease
, Inc. 8 Aug.
2003. <http://www.statease.com/articles.html>.
Milliken, William F., and Douglas L. Milliken. Race Car Vehicle
Dynamics. Warrendale, PA: Society of Automotive Engineers, 1995.
Montgomery, Douglas C. Design and Analysis of Experiments, 3
rd
edition. New
York: John Wiley & Sons, 1991.
Peng, K. C. The Design and Analysis of Scientific Experiments.
Reading, MA: Addison-Wesley Publishing Company, 1967.
Rusco, Mark S. Using Simulation Software for Design of Experiment
Training. Case Studies and Articles. Stat-Ease
, Inc. 8 Aug. 2003.

<http://www.statease.com/pubs/dragracing.pdf>.

Doe in Racing

Uploaded by

Copyright:

Available Formats

Doe in Racing

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Doe in Racing

Uploaded by

Copyright:

Available Formats

Design of Experiments in Racing

Design of Experiments in Racing

Software Users Guide. Version 6. Stat-Ease

, Inc. 8 Aug. 2003.

You might also like

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.