4. Product and Process Control Flashcards
The ability of a measurement method to replicate a given result
Precision
If lack of reproducibility is large in comparison to repeatability..
Operator training required
The response to the detection of a problem:
Containment
Diagnosis
Verification
Disposition
Reaction Plan
Used to identify critical parameters related to product performance and customer satisfaction
Product Features Matrix
Repeatability
Minimum variability in results
Repeatability Standard Deviation
Measure of the dispersion of the distribution of test results under repeatability conditions
Method to test the measurement process as well as ensure that the process yields data that is statistically stable
Measurement Systems Analysis
Fraction nonconforming that the customer is willing to accept some percentage of the time although it is considered by the customer to be reject quality
Limiting Quality
Design changes that results from failure traceability systems often…
Decrease warranty costs
Most non-conformities are defects in this test
Non-destructive test
Control of Standard Integrity
Controlling and preventing damage to measuring equipment
Amount of hand pressure used to hold a gauge is what type of error
Random Error
A calibration process is optimum when
Equipment transport is not a factor
Is 28% measurement error acceptable?
28% error is possibly acceptable, based on that particular application and equipment
Is 38% measurement error acceptable?
No - 38% error is unacceptable
Implies that the variability of the measuring instrument is consistent
Repeatability
Material Review Board roles
Review material that does not conform to standard
Determine what its disposition should be
Drive the development of effective corrective action to prevent recurrence
Critical control point identification (3)
- Impact on the customer
- Impact on the next step in the process
- Time required making corrections
Probability of acceptance vs fraction non-conforming - specific to each combination of sample size and acceptance number.
Operating Characteristic (OC) curves
Limit of a satisfactory process average
Acceptable Quality Limit (AQL)
The poorest quality in an individual lot that should be accepted
Lot Tolerance Percent Defective (LTPD/RQL/LQL)
The average quality after lots have been rejected (and then reworked) or accepted
Average Outgoing Quality (AOQ)
The worst possible average quality
Average Outgoing Quality Limit (AOQL)
The most expensive sampling plan
Single Sampling
Over a large number of samples, cheapest sampling plan
Double Sampling
This sampling plan require fewer samples if very good or very bad quality products
Multiple Sampling
Similar to multiple sampling but does not force a decision
Sequential Sampling
Can calculate OC curves and AOQ - taken off line during production
Continuous Sampling (CSP)
Designed to minimize average total inspection (ATI)
Dodge-Romig Sampling Tables
Sample Integrity
- An initial sample should be taken, documented, and maintained for traceability, where possible.
- The process conditions at the time of the initial sample should be recorded.
- All process references should be current and controlled
Initial samples are maintained as a reference and a confirmation
Sample Control
Accuracy and precision should be better than the preceding ones
Calibration
A fixed anvil and a spindle that moves lengthwise as it turns
Micrometer
Light rays are directed against the object and then reflected back through a projection lens onto a screen
Optical Comparators
Relies on acceptance/periodic sampling during the manufacture in order to prepare a control chart
Destructive Test
Non-Destructive Test examples
Liquid (dye) penetrant, magnetic particle testing, x-ray testing, ultrasonic testing, eddy current, gamma ray testing
Main objectives of Calibration (2)
- To reduce quality costs through the early detection of nonconforming products and processes with the use of measuring equipment of known accuracy.
- To provide customers (when they request it) with an indication of a supplier’s calibration capabilities.
A process designed to quantify a laboratory’s measurement uncertainty relative to the national standard. Takes into account bias and random error
Traceability
Possible results of measurement error
- Actions may be taken based on meaningless data.
- A process may appear to be out of control when it is not (and vice versa).
- Control problems may be seen after it is too late to find special causes.
- An experimental and mathematical method of determining how much the variation within the measurement process contributes to overall process variability
- Accuracy, Reproducibility, Repeatability
Measurement System Analysis (MSA)
Determining the variation of a measurement system:
- What approach should be used
- How many operators are to be involved
- How many sample parts are to be tested
- What number of repeat readings will be needed
Repeatability and Reproducibility (gage R&R) Studies
The minimum variability in results and implies that the variability of the measuring instrument itself is consistent
Repeatability
The maximum variability in results and implies that variability across operators is consistent
Reproducibility
The comparison of the measurement values between two or more different measurement systems
Measurement Correlation
Variation between a known standard across the low and high end of the gauge
Linearity
Using averages and range to study variability
- 2-3 replications (r)
- 2-3 appraisers (k)
- 10 parts (n)
Control Chart Method
Variation in accuracy readings over time on a given part
Stability
Accepted Reference Value – Average of Multiple Measurements
Accuracy
