PHASE I: PRELIMINARY STUDY INVESTIGATING MACHINE WARMUP AND SUPPLIERS PREVIOUS STUDIES


PHASE I: PRELIMINARY STUDY INVESTIGATING MACHINE WARMUP AND SUPPLIERS' PREVIOUS STUDIES

Description: This phase of the procedure is used to replicate and check the supplier's homework. It is used to verify machine warmup and setup, as well short- term stability.

Objectives: After completing this phase, the machine acceptance team should have verified the supplier's process setup and production methods , process warm-up period, and short-term (30 consecutive parts ) stability.

Phase I participants include the machine acceptance team and the supplier of the machine.

Method (Figure 16.8): Fundamentally there are three items of concern in this phase:

  1. Before data collection confirm gage capability by performing a repeatability and reproducibility (R&R) study (deemed capable if ratio < 0.1).

  2. Have vendor set up the process to make parts to your needs.

  3. Start up the process at production cycle times and wait the vendor-prescribed warmup period.

click to expand
Figure 16.8: Phase I flow chart.

Data collection: After supplier-prescribed warmup period, collect 30 consecutive parts. Number the parts to maintain the time order. Measure all 30 parts. Record the measurements in time order and inspect all parts for witness marks (deformations caused by handling). Have the supplier fix the machine if the witness marks are too frequent or destructive to the parts' performance.

Data analysis:

  • Plot the 30 measurements on an Xbar and R chart with the data arranged into subgroups of three. Look for any out-of-control conditions. Don't cheat yourself by being loose with the definitions of control. Remember that you are the one buying the machine.

  • Investigate the distribution of the 30 parts. If they appear to have a normal distribution, then proceed to calculate P pk . If the distribution is not normal, do calculations based on the mirror imaging technique or probability paper with nonnormal data.

  • If the control chart exhibits out-of-control conditions, then an action plan with the vendor must be developed and executed. After the problem has been identified and fixed, start back at the beginning of Phase I.

  • Potential capability for this size of data is determined to be four standard deviations within specification. If four standard deviations (99.994%) are in specification, then it is believed that there is a good chance of achieving long-term stability with three standard deviations (99.73%) within specification.

  • If C p and C pk are < 1.33, instruct the supplier to reduce the variation (reduce Rbar) in the process. If C p > 1.33 but C pk < 1.33, then the process needs to be centered on the middle of the specification. In either case, do not accept this machine or proceed to repeat Phase I study until after the supplier has made appropriate changes.

Proceed to phase II for a C pk > 1.33 or for a nonnormal distribution with 99.994% of its cumulative frequency distribution within specification. A good way to make sure that Phase I has met all the requirements is to use a checklist such as the one in Table 16.2.

Table 16.2: A Typical Checklist Before Moving to Phase II

If at least 99.994% of the distribution is not within the specification limits, then develop an action plan based on the following questions.

 

Yes

No

Action Required

Is the lack of capability because of the average?

_____

_____

____________

Is the lack of capability because of the range?

_____

_____

____________

Were there any trends or patterns over the time window of the study?

_____

_____

____________

Does the histogram suggest any unusual conditions such as nonnormality?

_____

_____

____________

Does the study log show any unusual occurrences that would help explain apparent incapability?

_____

_____

____________

Does either of the control charts give signals of unusual variation that would suggest stratification?

_____

_____

____________

Should the study be rerun?

_____

_____

____________

PHASE I EXAMPLE

Thirty parts were collected as part of a Phase I study. A histogram of the 30 parts is given. Note that the histogram is skewed to the low side, indicating process stream effects of warmup. The control chart does show a warmup trend as indicated by the histogram. From this, it is concluded that the supplier did not determine an adequate warmup period. The next Phase I study will wait 12 cycles (four groups of three) more than the supplier-determined warmup period before data are collected. Note that the R chart is in control; hence it makes sense to look at the Xbar chart.

Statistical (SQC) Data Analysis Phase I Example

Sample size

30

Actual

EST

Limits

 
     

99.730%

   

Target

15.0000

Low

14.5090

14.4339

14.5000

Average

14.9635

High

15.3036

15.4930

15.2000

Std. Dev.

0.1765

Range

0.7945

1.0591

0.7000

Skewness

-0.5175

% < Low limit

0.00

0.43

 

Kurtosis

0.0538

% > High limit

6.67

9.01

 

Normality test made

 

% Out of range

6.67

9.44

 

Normality assumed

         

Subgroup size = 3

         

WARNING: C p and C pk values for information only!

C P

0.6609

Process not stable due to: Xbar chart

C pk

0.4467

Cum Prob

Midpoint

Frequency

0.001

14.4000

0 +

0.002

14.4500

0 +

0.004

14.5000

1 + * ” ” ” ” ” ” ” ” ” ” ” ” ” ” ” ” ” ” ” ” ” ” ” ”

0.010

14.5500

0 +

0.020

14.6000

0 +

0.038

14.6500

2+.*

0.068

14.7000

0+.

0.113

14.7500

0+.

0.177

14.8000

1 + *.

0.260

14.8500

5 + **.**

0.360

14.9000

2 + **.

0.470

14.9500

3 + **.

0.582

15.0000

3 + **.

0.688

15.0500

6 + **. ***

0.780

15.1000

4 + **.*

0.855

15.1500

0+.

0.910

15.2000

1+. ” ” ” ” ” ” ” ” ” ” ” ” ” ” ” ” ” ” ” ” ” ” ” ”

0.948

15.2500

1 +.

0.972

15.3000

1 +.

0.986

15.3500

0 +

0.993

15.4000

0 +

0.997

15.4500

0 +

0.999

15.5000

0 +

1.000

15.5500

0 +

1.000

15.6000

0 +

Note: Histogram 1 "*" = 1 sample (actual data); "." = estimated distribution.

Statistical (SQC) Data Analysis Xbar Chart

LCL = 14.7087 Center = 14.9635 UCL = 15.2182

Subgroup size used: 3

Mean

SN

14.5000

14.7000

14.9000

15.1000

15.3000

   

+ ” ” ” ” + ” ” ” ” + ” ” ” ” + ” ” ” ” + ” ” ” ”

14.6564

 

I

*:

+

 

:

14.8760

 

I

:

*+

 

:

14.9894

 

I

:

+*

 

:

15.1273

 

I

:

+

*

:

15.0353

5

I

:

+

*

:

14.9598

 

I

:

+

 

:

14.9273

 

I

:

*+

 

:

15.0336

 

I

:

+

*

:

15.0184

 

I

:

+

*

:

15.0111

10

I

:

+

*

:

Range Chart

LCL = 0.000 Center = 0.2498 UCL = 0.6419

Subgroup size used: 3

Range

SN

0.0000

0.2000

0.4000

0.6000

0.8000

   

+ ” ” ” ” + ” ” ” ” + ” ” ” ” + ” ” ” ” + ” ” ” ”

0.3215

 

:

+

*

:

 

0.0457

 

: *

+

 

:

 

0.0225

 

: *

+

 

:

 

0.1617

 

:

* +

 

:

 

0.4722

5

:

+

*

:

 

0.5187

 

:

+

*

:

 

0.2688

 

:

+ *

 

:

 

0.1730

 

:

* +

 

:

 

0.2932

 

:

+ *

 

:

 

0.2202

10

:

* +

 

:

 



Six Sigma and Beyond. Statistical Process Control (Vol. 4)
Six Sigma and Beyond: Statistical Process Control, Volume IV
ISBN: 1574443135
EAN: 2147483647
Year: 2003
Pages: 181
Authors: D.H. Stamatis

flylib.com © 2008-2017.
If you may any questions please contact us: flylib@qtcs.net