Sei sulla pagina 1di 33

Operations

Management

Statistical Process Control

1
Statistical Process Control
(SPC)
Statistical technique used to ensure
process is making product to standard
All process are subject to variability
Natural causes: Random variations
Assignable causes: Correctable problems
Machine wear, unskilled workers, poor material
Objective: Identify assignable causes
Uses process control charts
2
Types of Statistical Quality Control
3
Statistical
Quality Control
Process
Control
Acceptance
Sampling
Variables
Charts
Attributes
Charts
Natural and Assignable Variation
4
Central Limit Theorem
X-charts based on this theorem
Regardless of distribution of population from
which samples are taken, the distribution of
sample means will tend to follow a normal
curve
Mean of sampling distribution will equal to mean of
population,
Std deviation of sample distribution will equal to
population std dev, o, divided by square root of
sample size, n
5
The Relationship Between Population and
Sampling Distributions
6
Uniform
Normal
Beta
Distribution of sample means
x means sample of Mean =
n
x
o
o = =
Standard deviation of
the sample means
(mean)
x
2 within fall x all of 95.5% o
x
3 within fall x all of 99.7% o

x
3
x
2
x
x
x
1
x
2
x
3 o + o + o 1 + o o o
Three population distributions
Control Chart Purposes
Show changes in data pattern
e.g., trends
Make corrections before process is out of
control
Show causes of changes in data
Assignable causes
Data outside control limits or trend in data
Natural causes
Random variations around average
7
Control Chart Types
8
Control
Charts
R
Chart
Variables
Charts
Attributes
Charts
X
Chart
P
Chart
C
Chart
Continuous
Numerical Data
Categorical or Discrete
Numerical Data
X Chart
Type of variables control chart
Shows sample means over time
Monitors process average
Example: Weigh samples of coffee &
compute means of samples; Plot
9
X Chart Control Limits
k
i
x
k
i 1
x
=

=
10
No of samples
Mean for
sample i
x
z x
x
LCL
x
z x
x
UCL
o
o
=
+ =
**When std dev, o, of population or process is available
x mean of sample means or target value set
z no of std dev as per confidence level
o
x
std dev of sample means
X Chart Control Limits
11
No of samples
Mean for
sample i
From
Table S6.1
R A x
x
LCL
R A x
x
UCL
2
=
2
+ =
k
i
x
k
i 1
x
=

=
**When std dev, o, of population or process is not available
Factors for Computing Control
Chart Limits (when o is not available)
12
Sample
Size, n
Mean
Factor, A
2

Upper
Range, D
4

Lower
Range, D
3

2 1.880 3.268 0
3 1.023 2.574 0
4 0.729 2.282 0
5 0.577 2.115 0
6 0.483 2.004 0
7 0.419 1.924 0.076
8 0.373 1.864 0.136
9 0.337 1.816 0.184
10 0.308 1.777 0.223
12 0.266 1.716 0.284

Control Chart for Samples of 9 Boxes
13
Variation due to
natural causes
17=UCL
16=Mean
15=LCL
Variation due to
assignable causes
Variation due to
assignable causes
Out of control
1 2 3 4 5 6 7 8 9 10 11 12
Sample Number
R - Charts
Type of variables control chart
Shows sample ranges over time, where
range is the diff between highest and
lowest values in each sample
Monitors process dispersion/variability
Example: How far from the mean are the
readings?
14
R Chart Control Limits
15
No of samples
Range for
sample i
From
Table S6.1
R D LCL
R D UCL
3
R
4
R
=
=
k
i
R
k
i 1
R
=

=
Steps to Follow When Using
Control Charts
16
1. Collect 20 to 25 samples of n=4 or n=5 from a
stable process and compute the mean.
2. Compute the overall means, set approximate
control limits (usually 3o),and calculate the
preliminary upper and lower control limits.If the
process is not currently stable, use the desired
mean (target) instead of the overall mean to
calculate limits.
3. Graph the sample means and ranges on their
respective control charts and determine
whether they fall outside the acceptable limits.
Steps to Follow When Using
Control Charts - continued
4. Investigate points or patterns that indicate
the process is out of control. Assign
causes for the variations.
5. Collect additional samples and revalidate
the control limits.
17
Mean and Range Charts
Complement Each Other
18
p Chart
Type of attributes control chart
e.g., good-bad
Shows % of nonconforming items
percent defective
Example: Count # of defective chairs &
divide by total chairs inspected; Plot
Chair is either defective or not defective
19
p Chart Control Limits
20
No of defective
Items in Sample i
Size of sample i
p = mean fraction defective
z = 2 for 95.5% limits;
z = 3 for 99.7% limits
sample each of size

where
n
x
p

z p LCL
z p UCL
i
k
1 i
i
k
1 i
p
p
=

=


=
=
+ =
=
=
n
n
p p
p
) 1 (
o
p
o
p
o
k = no of samples
c Chart
Type of attributes control chart
Shows number of nonconformities
(defects) in a unit
Unit may be chair, steel sheet, car etc.
Size of unit must be constant
Example: Count # defects (scratches,
chips etc.) in each chair of a sample of
100 chairs; Plot
21
c Chart Control Limits
22
# Defects in
Unit i
# Units Sampled
Use 3 for 99.7%
limits
k
c
c
i
k
1 i
=

=
=
+ =
c c LCL
c c UCL
c
c
3
3
Deciding Which Control Chart to
Use
Using an X and R chart:
Observations are variables
Collect 20-25 samples of n=4, or n=5, or more
each from a stable process and compute the
mean for the X chart and range for the R
chart.
Track samples of n observations each.
23
Deciding Which Control Chart to
Use
Using the P-Chart:
We deal with fraction, proportion, or percent
defectives
Observations are attributes that can be
categorized in two states
Have several samples, each with many
observations
Assume a binomial distribution unless the
number of samples is very large then
assume a normal distribution.
24
Deciding Which Control Chart to
Use
Using a C-Chart:
Observations are attributes whose defects per
unit of output can be counted
The number counted is often a small part of
the possible occurrences
Assume a Poisson distribution
Defects such as: number of blemishes on a
desk, number of typos in a page of text, flaws
in a square metre of cloth
25
Patterns to Look for in Control
Charts
26
What Is Acceptance Sampling?
Form of quality testing used for incoming
materials or finished goods
e.g., purchased material & components
Procedure
Take one or more samples at random from a
lot (shipment) of items
Inspect each of the items in the sample
Decide whether to reject the whole lot based
on the inspection results
27
What Is an Acceptance Plan?
Set of procedures for inspecting
incoming materials or finished goods
Identifies
Sample size (n)
Criteria (c) used to reject or accept a lot
Producer (supplier) & consumer (buyer)
must negotiate
Represented by Operating Characteristic
curve
28
Operating Characteristics Curve
Shows how well a sampling plan
discriminates between good & bad lots
(shipments)
Shows the relationship between the
probability of accepting a lot & its quality
29
OC Curve 100% Inspection
30
% Defective in Lot
P(Accept Whole Shipment)
100%
0%
Cut-Off
1 2 3 4 5 6 7 8 9 10 0
Return whole
shipment
Keep whole
shipment
An Operating Characteristic
(OC) Curve Showing Risks
31
o = 0.05 producers risk for AQL
|= 0.10
Consumers
risk for LTPD
Percent
Defective
Bad lots Indifference zone
Good
lots
LTPD AQL
0 1 2 3 4 5 6 7 8
100
95

75


50


25

10

0

Probability of
Acceptance
AQL & LTPD
Acceptable quality level (AQL)
Quality level of a good lot
Producer (supplier) does not want lots with
fewer defects than AQL rejected
Lot tolerance percent defective (LTPD)
Quality level of a bad lot
Consumer (buyer) does not want lots with
more defects than LTPD accepted
32
Producers & Consumers Risk
Producer's risk (o) Type I error
Probability of rejecting a good lot
Probability of rejecting a lot when fraction
defective is AQL or better; 5.05%
Consumer's risk () Type II error
Probability of accepting a bad lot
Probability of accepting a lot when fraction
defective is LTPD or worse; 10%
33

Potrebbero piacerti anche