Title: Statistical Process Control Chapter 4S
1Statistical Process ControlChapter 4S
- Learning Objectives
- Identify sources of variation
- Explain statistical process control
- Develop control charts for variables
- R chart,X chart
- Develop control charts for attributes
- P chart, c chart
- Explain acceptance sampling
- Producers consumers risk
2Statistical Quality Control (SPC)
- 1. Measures performance of a process
- 2. Uses mathematics (i.e., statistics)
- 3. Involves collecting, organizing,
interpreting data - 4. Objective Regulate product quality
- 5. Used to
- Control the process as products are produced
- Inspect samples of finished products
3Types of Statistical Quality Control
4Quality Characteristics
Attributes
Variables
- 1. Characteristics that you measure, e.g.,
weight, length - 2. May be in whole or in fractional numbers
- 3. Continuous random variables
- 1. Characteristics for which you focus on defects
- 2. Classify products as either good or bad,
or count defects - e.g., radio works or not
- 3. Categorical or discrete random variables
5Statistical Process Control (SPC)
- 1. Statistical technique used to ensure process
is making product to standard - 2. All process are subject to variability
- Natural causes Random variations
- Assignable causes Correctable problems
- Machine wear, unskilled workers, poor material
- 3. Objective Identify assignable causes
- 4. Uses process control charts
6Process Control Charts
7Control Chart Purposes
- 1. Show changes in data pattern
- e.g., trends
- Make corrections before process is out of control
- 2. Show causes of changes in data
- Assignable causes
- Data outside control limits or trend in data
- Natural causes
- Random variations around average
8Theoretical Basis of Control Charts
Central Limit Theorem
As sample size gets large enough (³ 30) ...
sampling distribution becomes almost normal
regardless of population distribution.
9Theoretical Basis of Control Charts
10Theoretical Basis of Control Charts
Properties of normal distribution
11Control Chart Types
R and X charts go hand in hand when monitoring
variables because they measure two critical
parameters 1. Central tendency
2. Dispersion
Continuous Numerical Data
Categorical or Discrete Numerical Data
Control
Charts
Variables
Attributes
Charts
Charts
R
P
C
X
Chart
Chart
Chart
Chart
12Statistical Process Control Steps
13X Chart
- 1. Type of variables control chart
- Interval or ratio scaled numerical data
- 2. Shows sample means over time
- 3. Monitors process average and tells whether
changes have occurred. These changes may due to - 1. Tool wear
- 2. Increase in temperature
- 3. Different method used
in the second shift - 4. New stronger material
- 4. Example Weigh samples of coffee compute
means of samples Plot
14X Chart Control Limits
From Table S4.1
Sample Range at Time i
Sample Mean at Time i
Samples
15R Chart
- 1. Type of variables control chart
- Interval or ratio scaled numerical data
- 2. Shows sample ranges over time
- Difference between smallest largest values in
inspection sample - 3. Monitors variability in process, it tells us
the loss or gain in dispersion. This change may
be due to - 1. Worn bearing
- 2. A loose tool
- 3. An erratic flow of
lubricant to machine - 4. Sloppiness of
machine operator - 4. Example Weigh samples of coffee compute
ranges of samples Plot
16R Chart Control Limits
From Table S4.1
Sample Range at Time i
Samples
17p Chart
- 1. Type of attributes control chart
- Nominally scaled categorical data
- e.g., good-bad
- 2. Shows of nonconforming items
- 3. Example Count defective chairs divide by
total chairs inspected Plot - Chair is either defective or not defective
18p Chart Control Limits
z 2 for 95.5 limits z 3 for 99.7 limits
Defective Items in Sample i
Size of sample i
19c Chart
- 1. Type of attributes control chart
- Discrete quantitative data
- 2. Shows number of nonconformities (defects) in a
unit - Unit may be chair, steel sheet, car etc.
- Size of unit must be constant
- 3. Example Count defects (scratches, chips
etc.) in each chair of a sample of 100 chairs
Plot
20c Chart Control Limits
Use 3 for 99.7 limits
Defects in Unit i
Units Sampled
21Process Capability Cpk
- Assumes that the process is
- under control
- normally distributed
If Cpk index goes above 1 the process becomes
more and more target oriented with fewer defects.
If Cpk is 1 the process variation is centered
within the control limits. However, if it is less
than 1, the process will not produce within the
specified limits
22What Is Acceptance Sampling?
- 1. Form of quality testing used for incoming
materials or finished goods - e.g., purchased material components
- 2. 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
23What Is an Acceptance Plan?
- 1. Set of procedures for inspecting incoming
materials or finished goods - 2. Identifies
- Type of sample
- Sample size (n)
- Criteria (c) used to reject or accept a lot
- 3. Producer (supplier) consumer (buyer) must
negotiate
24Operating Characteristics Curve
- 1. Shows how well a sampling plan discriminates
between good bad lots (shipments) - 2. Shows the relationship between the probability
of accepting a lot its quality
25OC Curve100 Inspection
P(Accept Whole Shipment)
100
Return whole shipment
Keep whole shipment
0
Cut-Off
Defective in Lot
26OC Curve with Sampling
P(Accept Whole Shipment)
Probability is not 100 Risk of keeping bad
shipment or returning good one.
100
Keep whole shipment
Return whole shipment
0
Cut-Off
Defective in Lot
27AQL LTPD
- 1. Acceptable quality level (AQL)
- Quality level of a good lot
- Producer (supplier) does not want lots with fewer
defects than AQL rejected - 2. Lot tolerance defective (LTPD)
- Quality level of a bad lot
- Consumer (buyer) does not want lots with more
defects than LTPD accepted
28Producers Consumers Risk
- 1. Producer's risk (a)
- Probability of rejecting a good lot (Type I
Error) - Probability of rejecting a lot when fraction
defective is AQL - 2. Consumer's risk (ß)
- Probability of accepting a bad lot (Type II
Error) - Probability of accepting a lot when fraction
defective is LTPD
29OC Curve Risk
30OC Curves for Different Sampling Plans
31Average Outgoing Quality
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 The maximum value of on the AOQ curve
correspond to the highest average percent
defective or the lowest average quality for the
plan It is called the Average Outgoing Quality
Limit (AOQL)
32Developing a Sample Plan
- 1. Negotiated between producer (supplier) and
consumer (buyer) - 2. Both parties attempt to minimize risk
- Affects sample size cut-off criterion
- 3. Methods
- MIL-STD-105D Tables
- Dodge-Romig Tables