with Complete Solutions
Why are samples taken? To determine whether assignable variations are present in the
output of a process
Central Limit Theorem Sample means will be normally distributed no matter whatever
distributions the samples follow
Types of Data -Variable: data that measures a particular product characteristic
(continuous) (length or width)
-Attribute: data that counts items (discrete) - number of defective items in a sample
Variable Charts - R chart
- X-bar chart
Attribute Charts -p-chart
-c-chart
,Fundamental concept in Control Charts -95.44% of all sample means fall within +- 2
standard deviations
-99.74% of all sample means fall within +- 3 standard deviations
Generic Control Charts -Have a center line that is the overall average
-Have limits above and below the center line at +- 3 standard deviations (usually)
Process centering - X bar chart
- X bar is a sample mean
- Size of each sample is n
Process Dispersion (consistency) - R chart
- R is sample mean
X Bar Charts -Center line is the grand mean (x double bar)
-Points are X bars
Range (R) Charts -Center line is the grand mean (R bar)
,-Points are R
Run tests -Even if a process appears to be in control, the data may not reflect a random
process
-Even if points are in the control limits the process must be random
C Charts -Used to count defects in samples of constant size
-Center line is the grand mean (c bar)
-Number of samples is m
P Charts -Used to track a proportion (fraction) defective
-Center line is the grand mean (p bar)
-Number of sample is m
-Size of each sample is n
Improving Process Capability -Simplify
-Standardize
-Mistake-proof
, -Upgrade equipment
-Automate
Centered Process -3 sigma: defects 2,700 parts per million
-6 sigma: defects 0.002 parts per million
Non-Centered Process -3 sigma: defects 67,000 parts per million
-6 sigma: defects 3.4 parts per million
Moment-of-truth analysis -experience enhancers
-standard expectations
-experience detractors
(order qualifier/order winner)
True or False: X-bar chart and R chart should be used together to find if a process is under
control. True