Basic Statistical Process

Control Training
By Carlos Sanchez
Content

 Quality Improvement & Statistics

 Variation
 What is Statistical Process Control?
 The Normal Distribution
 Let’s talk about Sigma
 SPC Tools
 Control Charts
 Process Capability

2
Why can’t we just take Samples?

Process Control Final Product Sampling

Inspection cost per unit is low Costs are Higher (Take sample, analyze, report,
dispose sample)
Inspection not destructive or detrimental to May be destructive or detrimental to our
our products products

Process can be adjusted, stopped, Process control is not feasible (After the fact) If
inspected and started up again at a product is out of spec now we have a full
reasonable cost tank/silo

3
But the product it’s in Spec!

 Meeting the specification is simply NOT ENOUGH, we need a

way to know this:
 How close to the target spec. is our product?
 How spread out were the results?

4
Variation

 Assignable Variation: Variation caused by factors that can be

clearly identified and possibly managed
Example: A poorly trained employee that
creates variation in finished product
output.

 Common Variation: Random variation which is caused by the

production process

Example: A particle classification process

that always allows bigger particles to flow
to the finished product

5
So, How do we improve?

 More Money  New & better equipment

 Flawless Raw Material
 Luck
 Reduce Common Variation…How?
 Observation, Observation…and more Observation
 Act on little changes observed
 Preventive Maintenance
 Statistical Process Control (SPC)
 Six Sigma
 Lean
 Design of Experiments

Continuous Improvement Tools

6
SPC in a nutshell

SPC quantifies variability and allows

you to determine if a process changed
 Minimize needless adjustments in the process (Tweaking)
 It’s a monitoring tool that lets us know when the process is
changing BEFORE product becomes UNACCEPTABLE/Out
Spec/ Unusable.
 It’s a prevention tool that allows to detect trends that could
lead to defective products. (Early warning system)
 Final inspection does not assure quality; remember: “You
can’t inspect quality into the product”
 Final Inspection is too late downstream

7
Two things to know about the Normal Distribution
LOCATION:
The Center of the
curve is
expressed as the
AVERAGE
This is where the
target Specification
is aimed at

SPREAD

SPREAD or RANGE:
The dispersion it is
usually expressed as
8
SIGMA
Let’s talk about Sigma
Sigma is just a fancy word for This is where the
Standard Deviation, which tells infamous SIX SIGMA
us how far is a particular value from comes from, it
the average of the data set. means sending
Product in spec.
99.73% of the
+/- 3 sigma
time
99.73%
+/- 2 sigma

96.45%

+/- 1 sigma
64.25%

9
Example

Imagine if an upside down bell curve could hold 100 Tons of Cement from a
storage silo.

If we are working
at +/- 1 sigma
only 64 Tons are
in Spec.

If we are working
at +/- 3 sigma
If we are working
at +/- 2 sigma 64 Tons almost all 100 Tons
are in Spec.
only 96 Tons are
in Spec.
96 Tons

99.7 Tons
10
7 Classical SPC Tools
Control Chart Histogram
Pareto Chart

Cause Effect Chart

Stratification Chart

Flow Chart

Check Sheet

For this initial

Training we will focus on:
Control Charts

11
Why use Control Charts?

 Reduce variation by the systematic elimination of assignable

causes
 Prevent unnecessary process adjustments (Tweaking)
 Visually diagnose the process by observing data patterns
 Find out what our process can do
 Provide immediate visual feedback
 Decide if continuing production is worthwhile

12
Types of Control Charts

 Run Charts for variable data:

 Individual Chart
 Mean & Range Charts We will focus on these today
 Std. Dev. Charts
 Attribute Charts

13
Why don’t we just use the Specs. As our Limit?
Too late…It’s bad

Upper Spec.

Target

Lower Spec.

With limits we have a “cushion or safety net”

before the S#$@%! Hits the fan!

14
So where should these Control Limits be?

Where would you put a Control Limit?

+/- 3 sigma
99.73%
+/- 2 sigma

96.45%

+/- 1 sigma
64.25%

15
How is a chart related to the Normal Curve?

Upper Spec.
Upper Control Limit

Lower Control Limit

Lower Spec.

16
Let’s tilt the Chart and let the points fall!

Huh!
That makes
sense!

17
At the end it averages out!

 When the population is big, looking at individuals to detect

trends is tricky…
 It’s been proven that when you look at averages these tend
to behave like a Normal Curve
 Google this: Central Limit Theorem (It’s great for those sleepless nights)

 So from now on this training all example charts are based on

Averages. This means that a “Point” in a control charts
represents the “Average” value of a sample (Typical sample
size varies from 3 to 5), I like 5, but heck, you can choose
whatever size you want

18
Usefulness of looking at Average & Range

(process mean is
shifting upward)
Sampling
Distribution

UCL

x-Chart Detects shift

LCL

UCL

Does not
R-chart
detect shift
LCL

19
More Usefulness of looking at Average & Range

Sampling (process variability

Distribution is increasing)

UCL

x-Chart Does not

LCL Detects shift
UCL

Detect shift
R-chart
LCL

20
If you really want to plot a chart by hand…Ok!
Average of all Averages
Average Range
Sample
Constants
x Chart Control Limits Size

n A2 D3 D4
UCL = x + A 2 R 2 1.88 0 3.27
3 1.02 0 2.57
LCL = x - A 2 R 4 0.73 0 2.28
5 0.58 0 2.11
6 0.48 0 2.00
R Chart Control Limits 7 0.42 0.08 1.92
8 0.37 0.14 1.86
UCL = D 4 R 9 0.34 0.18 1.82
10 0.31 0.22 1.78
LCL = D 3 R 11 0.29 0.26 1.74

21
Let’s Analyze that Chart

Points out of Control Limits : Rule of thumb, if there are any point outside the
Control limits should be investigated.

Upper Spec.
Upper Control Limit

Lower Control Limit

Lower Spec.

22
Let’s Analyze that Chart

Trends : Rule of thumb, if there are 7+ points in a row all higher or lower
than the preceding point

Upper Spec.
Upper Control Limit

Lower Control Limit

Lower Spec.

23
Let’s Analyze that Chart

Shifts : Rule of thumb, if there are 5+ points in a row all higher or lower
than target or Average, this means that the Average has SHIFTED

Upper Spec.
Upper Control Limit

Lower Control Limit

Lower Spec.

24
Let’s Analyze that Chart

Cycle : Rule of thumb, if there are 3+ similar peaks or valleys, this is typical of
Machine wear, or dosage cycles…or Tweaking!

Upper Spec.
Upper Control Limit

Lower Control Limit

Lower Spec.

25
Let’s Analyze that Chart
Adherence to Center : Rule of thumb, if there are 7+ all smothering the average
or target spec. This means that the measurement equipment is no longer capable
of detecting significant variation. This is good, but it signals for improvement in the
measurement system. Maybe the spec can be tightened.
Upper Spec.
Upper Control Limit

Lower Control Limit

Lower Spec.

26
Let’s Analyze that Chart
Erratic : Rule of thumb, if there are 6+ points shifting from one extreme of the
chart to the other, borderline with the Control Limits, this shows that the process
is not stable…When you see this pattern be alert for Non conforming product.
Upper Spec.
Upper Control Limit

Lower Control Limit

Lower Spec.

27
Let’s talk about Adjustment or Tweaking the Process
If your process is not capable, then there is a good chance that some of your
sample will have values outside the specification. Chances are if you are not
looking at a SPC control chart, you may be tempted to make an adjustment.
Let's see what would happen.

x
x x
xx
x x
xx
x xx
x
x x
xxxxxxxxx x
xx xxxxxxxx
x
x x x xxxxxxxxxx x x x
x x xxxxxxxxxx
x x
x x
x x
xx
xx
xx
xx
xx
xx xxx x x
x x x xxxxxx
x x x xxxxxxx xx
x xx
x x xxx xxxxxxx xx
xx
xxxxxxxxxxxx x xx xxxxxxxxx xx
xxxxx xx
xx
xx
xxxxx
xxxx x
x
x
x x
xxxxxxxxx
x xxx x xx
xx
xx
xxxxxxxx
xx
xxx
xxxx
xxxx
xx x x
x x x
xx
xx
xx
xx
xxxxxx
x xxxx
xx xxx
x xxx x
x x
x x
x x
xx
xx
xx
xx
xx
xx
x x
xx
x
xx x xxx
xx xxxxxxxxxxxx
xxxxx xxxxxxx x x
xx xxxx
xxxx x xx
xxxx
xx
x xxx
xx
xx

28
Let’s try to understand Process Capability

Jack, Nice guy; works as a technician. He lives 10 miles Jim. Nice guy too, Works in Plant B as a research
away from the Plant A. In order to get to work he takes assistant, he lives 5 miles from work. In order to get
Hwy. 7 and gets off at the Beachtree exit and zips right to work he has to get through 5 traffic light onto Hwy
into work. He never hits any traffic and there is no traffic 4 (which is frequently backed up by crazy skiers) to
light between his home and work. downtown Beachtree. There he has to find parking
He's never late to work. spot, sometimes a couple of blocks away.
He is late to work quite frequently.

29
Process Capability Continued
Jack Jim
Arrives to work between Arrives to work between
7:48 to 7:56 AM. 7:48 to 8:06 AM

Early
to Late to work
work

Tolerance

7:42 7:48 7:54 8:00 8:06 8:12

If we thought of being early or late to work as our specification, then we

can say that Jack is Capable meeting the specification. Jim is Not
Capable of meeting the specification.
30
Let’s Calculate their Capability to get on Time
Jack Jim
Arrives to work between Arrives to work between
7:48 to 7:56 AM 99.7% of time. 7:48 to 8:06 AM 99.7% of time.
6 sigma = 7:56 -7:48 = 8 min. 6 sigma = 8:06 - 7:48 = 18 min.

Tolerance = late - early

Tolerance = 8:00 - 7:46
Tolerance = 14 minutes
Capability = Tolerance
6 sigma
If Capability is > 1 then we can conclude
that is capable of meeting the spec

Jack's Capability = 14 / 8 Jim's Capability = 14/18

= 1.75 (Bill is capable) = 0.78 (Jim is NOT capable)

31
Ok, lets get to Cp & Cpk…Say what?
Real
Avg.
Target

Out Out
Spec Spec

Tolerance

Cp: Measures the capability Cpk: Measures the capability

of the process to meet the of the process to meet the Target Spec.
Tolerance…just like Jack & Jim it looks at the likelihood of making
product out spec. So the more
“centered” the curve is, a better cpk
you will get.
32
More formulas…But they are short!

Cp = Upper Spec. – Lower Spec.

6 sigma

Cpk = The smallest of:

Target - lower spec or Upper spec - Target

3 sigma 3 sigma

Criteria:
Both Cp & Cpk should be AT LEAST > 1
Ideally > 1.33
Why? Just trust me on this one…

33
34