Search
Create
Log in
Sign up
Log in
Sign up
Control Charts
STUDY
Flashcards
Learn
Write
Spell
Test
PLAY
Match
Gravity
Terms in this set (21)
Steps in Design of Quality Control Systems
Identify critical points
Incoming materials & services
Work in process
Finished product or service
Decide on the type of measurement
variables
attribute
Decide on the amount of inspection to be used.
Decide who should do the inspection
Statistical Process Control terms:
Target value
Specification limits---Range of design acceptance
Control limits---Range of process stability
Process variability is inevitable
Process Capability Index
ratio < 1 = not sigma capable
On target both numbers are the same
(use when off target)
Process Capability Ratio
The larger the ratio the more capable the process:
ratio > 1 means 97.73 of output is in specifications
(only use if on target)
Statistical Process Control (SPC)
n
Statistical Quality Control
Basic assumptions (tenets) of Statistical Quality Control:
- Every process has random variation in it.
- Production processes are not usually found in a state of control.
"State of Control"; what does it mean?
- Unnecessary variation is eliminated.
- Remaining variation is because of random causes
Statistical Sampling
100% inspection is costly ... and unnecessary ...
Sometimes is the only way to go (destructive tests)
Good sampling allows conclusions about population
Sample means follow normal distribution (Central Limit Theorem)
Six Sigma Quality
Six Sigma uses a project/team approach.
A process is selected for improvement
A cross-functional team is formed.
A six sigma 'black belt' is chosen to head the team.
The team uses the DMAIC (Define; Measure; Analyze; Improve; Control) method for finding root causes and improving the process.
Taguchi Techniques
Loss Function.
-Experimental design methods to improve product & process design
- Pro: Identify key component & process variables affecting product variation with small samples
- Con: Ignore interaction effect among variables
Taguchi Concepts
- Quality robustness: remove effects not causes
- Quality loss function
- Target specifications
When find a defect:
Containment: Keep the defective items from getting to the customer
Correction: Find the cause of the defect and correct it.
Prevention: Prevent the cause from happening again.
Continuously monitor & improve the system.
Type of Quality Characteristic
Variable Data: (x-bar & R chart)
can be measured on a continuous scale
e.g. size, weight, volume
Attribute Data: (P or C chart)
counts, such as the number (or proportion) of defects in a sample
the presence or absence of an attribute
e.g. smooth finish, light turns on (works)
X-bar Chart
Type of variables control chart
Interval or ratio scaled numerical data
Shows sample means over time
Monitors process average
Example
Weigh samples of boxes of oat flakes
Compute means of samples
Plot
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
Control Charts for attribute Data
Measuring # defective vs. # of defects
Attribute data control charts:
P-Chart for the fraction defective
C-Chart for the number of defects (Poisson distr.)
P -Chart
Type of attributes control chart
Nominally scaled categorical data
e.g., good-bad
Shows % (proportion) of nonconforming items
Example:
Count # defective chairs
Divide by total chairs inspected
Plot
Note that chair is either defective or not defective
C Chart
Type of attributes control chart
Discrete quantitative data
Assumes Poisson Distribution
Shows number (count) 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
Inspection:
Quality should be designed into the product or service process. However, inspection may still be needed. The question is just where to inspect?
Inspection stations are generally found:
- at raw materials inputs.
- at costly operations.
- at bottleneck operations.
- at finished goods.
The number and location of inspection stations must be balanced against the cost of passing defective materials to the next step.
Acceptance sampling
Accept= consumer Risk
Reject = Producers Risk
Lean Sigma
Latest Buzzword
Lean & Six Sigma are complementary approaches to improvement.
Lean seeks to eliminate waste.
Six sigma seeks to eliminate defects.
Six sigma organization is more formal and training intensive.
Six sigma is project focused; lean is more broad based.
Difference Between Process Capability and Statistical Process Control
Process capability is something you check before you start producing.
Statistical process control (SPC) is something you run continually during production.
Sources of Variability
Common Causes (Natural Variation)
frequent
short term
random
Special Causes (Assignable Variation)
less frequent (hopefully)
identifiable (hopefully)
preventable (hopefully)
;