Robust Design of an Automotive Suspension System
Sreeram Parameswaran | MS | 1994
Robust Design of an Automotive Suspension System
Sreeram Parameswaran | MS | 1994
In the past 20 years the American industry has lost its competitive
edge in manufacturing. A significant portion of this loss has been attributed
to a decline in product quality. From a technical standpoint this may be
due to a lack of effective implementation of Statistical Process Control
(SPC) and Statistical Experimental Design (SED) methods to improve quality
and productivity. Ironically, both SPC and SED were mainly developed in
the western hemisphere but Japan was able to effectively apply these methods
to improve their competitiveness. The success of the Japanese gave strong
thrust in the direction of statistical quality control to improve design.
One of the more popular methods in the new Quality philosophy was introduced
by Taguchi and combines some elements of classical design techniques with
cost considerations. Some methods like the Taguchi Loss function approach
to quality improvement were designed to help find less costly solutions
while maintaining high quality and productivity standards. Other contemporary
philosophies like the Response Surface Methodology (RSM) have also emerged
as strong players. These tools of quality improvement are used in a number
of applications throughout the engineering world and some of these are
used in this research.
The kinematic design analysis of a front suspension system is one such
field where these statistical quality improvement tools can be used in
conjunction with the field of kinematics to provide tangible improvements
in the performance of an automobile. For the purpose of understanding and
implementing these statistical quality control tools, a kinematic model
of a McPherson strut suspension system was developed. The performance of
such a system, which is a function of the lengths and orientations of its
links, may be sensitive to their manufacturing and assembly variations
which may alter its design intent. Recognizing this fact and trying to
produce a robust product that is less sensitive to these variations, is
attempted in this research.
The loss function is an attractive measure of any system. This function jointly considers the mean and variation of the product performance. When it approaches zero we have a product that makes good monetary sense. The scrub is one of the measures of performance of the suspension system. To model the loss as a function of scrub in an attempt to push the loss towards its lowest absolute limits possible, and hence obtain the most commercially viable suspension design, is the crux of this research.
If you have any comments or suggestions please
e-mail jwsuther@mtu.edu.