14.10.2015 Views

100-Metodos-de-Qualidade-Total

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

DATA COLLECTION, ANALYSIS AND DISPLAY<br />

233<br />

Table 1 Data used to monilor thickness of machined parts<br />

Lot no.<br />

2 3 4 5<br />

Xl X2 XJ X4 X;<br />

Sum<br />

X<br />

Mean<br />

X<br />

Range<br />

R<br />

2<br />

3<br />

4<br />

5<br />

6<br />

21.0 + 18.9 + 19.8 + 19.7 + 18.2<br />

19.8 + 19.9 + 19.1 + 20.0 + 18.2<br />

20.0 + 19.2 + 19.5 + 20.2 + 18.6<br />

19.5 + 18.6 + 19.9 + 19.7 + 19.8<br />

20.5 + 18.0 + 18.8 + 18.6 + 18.6<br />

20.8 + 18.1 + 19.0 + 20. 1 + 19.5<br />

97.6<br />

97.0<br />

97.5<br />

97.5<br />

94.5<br />

97.5<br />

19.52<br />

19.40<br />

19.50<br />

19.50<br />

18.90<br />

19.50<br />

2.8<br />

1.8<br />

1.6<br />

1.3<br />

2.5<br />

2.7<br />

29<br />

30<br />

18.1 + 20.2 + 20.2 + 18.4 + 18.8<br />

20.9 + 19.2 + 18.0 + 18.4 + 18.3<br />

95.7<br />

94.8<br />

19.14<br />

18.96<br />

2.1<br />

2.9<br />

<strong>Total</strong><br />

567.0<br />

57.0<br />

Mean<br />

18.9<br />

1.9<br />

Benefit<br />

It can be difficult to separate out random variation (often called common<br />

cause or non-assignable variation) from real variation caused by changes to<br />

the process. X-R charts are a way of doing this for the variables data with a<br />

sample size that is greater than 1.<br />

Example<br />

Figures 1 and 2 show the use of an X-R chart to monitor the thickness of<br />

machined parts using the data given in Table 1. The number of sampling<br />

points m = 30; the size of each sample n = 5.<br />

Reference<br />

m = 30n = 5<br />

'iR 57<br />

R =<br />

= = 1 . 9<br />

m 30<br />

'iX 567<br />

X = = = 18 . 9<br />

m 30<br />

UCLR = D4R = 2.11 X 1.9 = 4<br />

LCLR = D3R = 0 x 1. 9 = 0<br />

A2R_ = 0.58 x 1.9 = 1.1<br />

UCLx = _X+ A2R = 18.9 + 1.1 = 20<br />

LCLx = X-A2R = 18.9 - 1.1 = 17.8<br />

J.S. Oakland and R.F. Followell (1994) Statistical Process Control. London: Butterworth/<br />

Heinemann.

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!