Statistical Process Control

Outline
 STATISTICAL PROCESS CONTROL (SPC)
       Control Charts for Variables (x - Charts) The Central Limit Theorem Setting Mean Chart Limits Setting Range Chart Limits (R-Charts) Using Mean and Range Charts Control Charts for Attributes Managerial Issues and Control Charts

Outline
 PROCESS CAPABILITY
 Process Capability Ratio (Cp)  Process Capability Index (Cpk

 ACCEPTANCE SAMPLING
 Operating Characteristic (OC) Curves  Average Outgoing Quality

Learning Objectives
When you complete this chapter, you should be able to Identify or Define:
        Natural and assignable causes of variation Central limit theorem Attribute and variable inspection Process control charts x-charts and R LCL and UCL p-charts pk C and C and C-charts
p
4

Learning Objectives Continued

When you complete this chapter, you should be able to Identify or Define:
     Acceptance sampling OC curve AQL and LTPD AOQ Producers and consumers risk

Learning Objectives Continued

When you complete this chapter, you should be able to Describe or explain:
 The role of statistical quality control

Statistical Quality Control (SPC)

 Measures performance of a process  Uses mathematics (i.e., statistics)  Involves collecting, organizing, & interpreting data  Objective: provide statistical signal when assignable causes of variation are present  Used to
 Control the process as products are produced  Inspect samples of finished products

Types of Statistical Quality Control

Statistical Quality Control

Process Control

Acceptance Sampling

Variables Charts

Attributes Charts

Natural and Assignable Variation

Quality Characteristics
Variables
Characteristics that you measure, e.g., weight, length May be in whole or in fractional numbers Continuous random variables

Attributes  Characteristics for which you focus on defects  Classify products as either good or bad, or count # defects
 e.g., radio works or not

 Categorical or discrete random variables

10

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

11

Process Control: Three Types of Process Outputs

Frequency Lower control limit
(a) In statistical control and capable of producing within control limits. A process with only natural causes of variation and capable of producing within the specified control limits. Upper control limit (b) In statistical control, but not capable of producing within control limits. A process in control (only natural causes of variation are present) but not capable of producing within the specified control limits; and (c) Out of control. A process out of control having assignable causes of variation.
12

Size (Weight, length, speed, etc. )

The Relationship Between Population and Sampling Distributions

Three population distributions Distribution of sample means
Beta

Normal

Mean of sample means ! x Wx Standard deviation of ! Wx ! the sample means n

Uniform  3W x  2 W x  1W x  W x  2 W x  3W x

x (mean)

95.5% of all x fall within s 2 W x 99.7% of all x fall within s 3W x

13

Sampling Distribution of Means, and Process Distribution

Sampling distribution of the means
Process distribution of the sample

x!m ( mean )
14

Process Control Charts

Plot of Sample Data Over Time
80 Sample Value 60 40
Average Sample Value UCL

20
LCL

0 1 5 9 13 Time
15

17

21

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
16

Theoretical Basis of Control Charts

Central Limit Theorem
As sample size gets large enough, sampling distribution becomes almost normal regardless of population distribution.

X
X
17

Theoretical Basis of Control Charts

Central Limit Theorem
Mean Standard deviation
Wx Wx ! n

X !Q

X
X !Q
18

Theoretical Basis of Control Charts

Properties of normal distribution
95.5% of all x fall within s 2W x 99.7% of all x fall within s 3W x

x !Q
19

Control Chart Types

Continuous Numerical Data Control Charts Categorical or Discrete Numerical Data

Variables Charts

Attributes Charts

R Chart

X Chart

P Chart

C Chart
20

Statistical Process Control Steps

Start Produce Good Provide Service Take Sample No
Can we assign causes?

Yes Inspect Sample Create Control Chart Stop Process Find Out Why

21

DX Chart
 Type of variables control chart
 Interval or ratio scaled numerical data

 Shows sample means over time  Monitors process average  Example: Weigh samples of coffee & compute means of samples; Plot

22

Control Chart for Samples of 9 Boxes

Variation due to assignable causes 17=UCL 16=Mean 15=LCL Variation due to assignable causes Variation due to natural causes

1 2

3 4

8 9 10 11 12

Sample Number

Out of control

23

DX Chart Control Limits

UCLx ! x  A R LCLx ! x  A R
Range for sample i From Table S6.1

Mean for sample i

x !

i !

xi n
R !
# Samples

i !1

Ri n
24

Factors for Computing Control Chart Limits

Sample Size, n 2 3 4 5 6 7 8 9 10 12 Mean Upper Lower Factor, A2 Range, D4 Range, D3 1.880 3.268 0 1.023 0.729 0.577 0.483 0.419 0.373 0.337 0.308 0.266 2.574 2.282 2.115 2.004 1.924 1.864 1.816 1.777 1.716 0 0 0 0 0.076 0.136 0.184 0.223 0.284
0.18 4

25

R Chart
 Type of variables control chart
 Interval or ratio scaled numerical data

 Shows sample ranges over time

 Difference between smallest & largest values in inspection sample

 Monitors variability in process  Example: Weigh samples of coffee & compute ranges of samples; Plot
26

R Chart Control Limits

UCL R ! D 4 R
From Table S6.1

LCL R ! D3R

R ! i !1 n

Ri

Range for Sample i

# Samples
27

Steps to Follow When Using Control Charts

2. Compute the overall means, set approximate control limits,and calculate the preliminary upper and lower control limits.If the process is not currently stable, use the desired mean 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.

1. Collect 20 to 25 samples of n=4 or n=5 from a stable process and compute the mean.

28

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.

29

Mean and Range Charts Complement Each Other

30

p Chart
 Type of attributes control chart
 Nominally scaled categorical data
 e.g., good-bad

 Shows % of nonconforming items  Example: Count # defective chairs & divide by total chairs inspected; Plot
 Chair is either defective or not defective
31

p Chart Control Limits

UCL p ! p  z LCL p ! p  z ni
i !1 k

p (1  p ) n

p (1  p ) n xi ni
i !1 i !1 k k

z = 2 for 95.5% limits; z = 3 for 99.7% limits

n!

and p !

# Defective Items in Sample i Size of sample i

p (1  p ) Wp ! n where n ! size of each sample

32

P-Chart for Data Entry Example

0.12 0.11 0.10 0.09 0.08 0.07 0.06 0.05 0.04 0.03 0.02 0.01 0.00 Sample Number
33

UCLp=0.10

P e r c e n t D e fe c tiv e

p ! 0.04

LCLp=0.00
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

c Chart
 Type of attributes control chart
 Discrete quantitative data

 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

34

c Chart Control Limits

UCLc ! c  3 c LCLc ! c  3 c ci
i!1 k

Use 3 for 99.7% limits

# Defects in Unit i # Units Sampled

35

c !

Patterns to Look for in Control Charts

36

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.

 Using the P-Chart:

 We deal with fraction, proportion, or percent defectives  Observations are attributes that can be categorized in two states

37

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 bolt of cloth

38

Process Capability Ratio, Cp

Upper Specification  Lower Specification Cp ! 6 W ! standard deviation of the process

39

Process Capability Cpk

Upper Specification Limit  x C pk ! minimum of , or 3W x  Lower Specification Limit 3W where x ! process mean W ! standard deviation of the process population
Assumes that the process is:
under control normally distributed

40

Meanings of Cpk Measures

Cpk = negative number Cpk = zero Cpk = between 0 and 1 Cpk = 1 Cpk > 1

41

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
42

What Is an Acceptance Plan?

 Set of procedures for inspecting incoming materials or finished goods  Identifies
 Type of sample  Sample size (n)  Criteria (c) used to reject or accept a lot

 Producer (supplier) & consumer (buyer) must negotiate

43

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

44

OC Curve 100% Inspection

P(Accept Whole Shipment) 100%
Keep whole shipment Return whole shipment

0%

Cut-Off

7 8 9 10 % Defective in Lot
45

OC Curve with Less than 100% Sampling

P(Accept Whole Shipment)
100% Probability is not 100%: Risk of keeping bad shipment or returning good one.

Keep whole shipment

Return whole shipment 0%

3 4 Cut-Off

10
46

% Defective in Lot

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
47

Producers & Consumers Risk

 Producer's risk (E)
 Probability of rejecting a good lot  Probability of rejecting a lot when fraction defective is AQL

 Consumer's risk ()
 Probability of accepting a bad lot  Probability of accepting a lot when fraction defective is LTPD
48

An Operating Characteristic (OC) Curve Showing Risks

100 95 75

E = 0.05 producers risk for AQL

Probability of Acceptance

50

25 10

F= 0.10
Consumers risk for LTPD 0 0 1 Good lots 2 AQL Indifference zone 3 4 5 6 7 LTPD Bad lots 8

Percent Defective

49

OC Curves for Different Sampling Plans

P(Accept Whole Shipment) 100% n = 50, c = 1 n = 100, c = 2

0%

3 4 5 6 7 8 AQL LTPD % Defective in Lot

10

50

Average Outgoing Quality

AOQ ! ( Pd )( Pa )( N  n ) N

Where:

Pd = true percent defective of the lot Pa = probability of accepting the lot N = number of items in the lot n = number of items in the sample

51

Developing a Sample Plan

 Negotiate between producer (supplier) and consumer (buyer)  Both parties attempt to minimize risk
 Affects sample size & cut-off criterion

 Methods
 MIL-STD-105D Tables  Dodge-Romig Tables  Statistical Formulas
52

Statistical Process Control Identify and Reduce Process Variability

Lower specification limit Upper specification limit (a) Acceptance sampling [ Some bad units accepted; the lot is good or bad] (b) Statistical process control [Keep the process in control] (c) cpk >1 [Design a process that is in control]

53