Quality in Design - Medical Device Design: Innovation from Concept to Market - page 124

Table 7.3: Experiment Design for a Four-Variable System

Experiment Variable 1 (X1) Variable 2 (X2) Variable 3 (X3) Variable 4 (X4)

1

+ 1 + 1 + 1 + 1

2

+ 1 + 1 + 1 − 1

3

+ 1 + 1 − 1 + 1

4

+ 1 + 1 − 1 − 1

5

+ 1 − 1 + 1 + 1

6

+ 1 − 1 + 1 − 1

7

+ 1 − 1 − 1 + 1

8

+ 1 − 1 − 1 − 1

9

− 1 + 1 + 1 + 1

10

− 1 + 1 + 1 − 1

11

− 1 + 1 − 1 + 1

12

− 1 + 1 − 1 − 1

13

− 1 − 1 + 1 + 1

14

− 1 − 1 + 1 − 1

15

− 1 − 1 − 1 + 1

16

− 1 − 1 − 1 − 1

Table 7.4: Experiment Design with Settings

Experiment Variable 1 (X1) Variable 2 (X2) Variable 3 (X3) T P Tm

1 + 1 + 1 + 1 Max Max Max

2 + 1 + 1 − 1 Max Max Min

3 + 1 − 1 + 1 Max Min Max

4 + 1 − 1 − 1 Max Min Min

5 − 1 + 1 + 1 Min Max Max

6 − 1 + 1 − 1 Min Max Min

7 − 1 − 1 + 1 Min Min Max

8 − 1 − 1 − 1 Min Min Min

Now we come to the interesting bit, we do the experiments. Using the table we add the
settings values; they could be just Max and Min, or they could be actual numbers, it matters
not ( Table 7.4 ). We have eight experiments to conduct, but we must measure something.
We could measure quantitative numbers such as weight, surface finish or number of
surface defects. Equally we could be quite subjective and grade how it feels or looks (from
1-10 say). Again it matters not so long as we have a measure of quality described by a
number. Before we do the experiment we mess up the data using a random selection for the
experiment order; this eliminates any order effects. The final column is the results from the
experiment (in Table 7.5 perceived quality is rated from 0-5 where 5 is excellent and 0 is
awful). It is a really good idea to get a third party to do the experiments. If you do this you
will have just made the test more statistically relevant by removing your influence from the
results.