Page 571 - Design for Six Sigma a Roadmap for Product Development
P. 571
Design Optimization:Taguchi’s Robust Parameter Design 529
4. Select adequate inner and outer arrays; assign control factors to the
inner array and noise factors to the outer array.
5. Perform the experiment.
6. Perform statistical analysis based on S/N ratio; sometimes it is
helpful to analyze the interactions between control factors and
noise factors.
7. Predict optimal output performance level based on optimal control
factor level combination, and conduct a confirmation experiment to
verify the result.
Example 14.6 illustrates these steps.
Example 14.6. Maximize Pulloff Force for an Elastometric Connector to a
Nylon Tube This Taguchi parameter design project involves finding the
optimal process parameters in assembling an elastometric connector to a
nylon tube in an automobile engine application. It is required that the
strength of the connection be strong. The quality characteristic measured in
this project is the pulloff force, which is a larger-the-better characteristic.
There are four control factors in this experiment:
Control factors Level 1 Level 2 Level 3
A Interference Low Medium High
B Connector wall thickness Thin Medium Thick
C Insertion depth Shallow Medium Deep
D Percent adhesive in Low Medium High
connector predip
There are three noise factors as follows:
Noise factors Level 1 Level 2
U Conditioning time, h 24 120
V Conditioning temperature, °F 72 150
W Conditioning relative humidity, % 25 75
In this experiment two compound noise factors are selected as follows:
N1 Set conditioning time and conditioning temperature at low level, and
set humidity at high level, that is, U and V at level 1, and W at level 2; this
noise combination will weaken the connection.
N2 Set conditioning time and conditioning temperature at high level and
humidity at low level, that is, U and V at level 2 and W at level 1; this noise
combination will strengthen the connection.
An L 9 array is used as the inner array. Pulloff forces are measured for each
inner array run with replicates at two compounded noise combinations.
The experimental layout and experimental data are listed in the following
table:
