Frito-Lay Statistical Process Control HAROLD COBURN AND MACK BURT - - PowerPoint PPT Presentation
Frito-Lay Statistical Process Control HAROLD COBURN AND MACK BURT Statistical Process Control The Application of statistical Techniques to control production process. Acceptance sampling is used to determine acceptance or rejection
HAROLD COBURN AND MACK BURT
The Application of statistical Techniques to control production process.
Acceptance sampling is used to determine acceptance
S.P.C is widely used to ensure process meets standards.
All processes are subject to a certain degree of variability.
Affect almost every production process and
are to be expected.
Are many sources of variation that occur
within a process that is in statistical control.
When grouped together they form a pattern
described as distribution.
When normal the are characterized by two
parameters.
Mean µ-The measure of central tendency. Standard deviation Ơ- The measure of
dispersion.
Can be traced to a specific reason.
Factors such as machine wear, misadjusted equipment, fatigued
Assignable and natural variations distinguish two task for operation managers.
Task 1 is ensuring that the process can operate under control.
Task 2 identify and eliminate assignable variations.
Major role in SPC. Proper sampling can allow you to discover any
issue that are hidden.
Charts are you used to visually see the whether or
not the product exist between the control parameters.
X chart monitors the arithmetic means of successive samples
characteristics that can be measured on a continuous scale, such as weight, temperature, thickness etc. For example, one might take a sample of 5 shafts from production every hour, measure the diameter of each, and then plot, for each sample, the average of the five diameter values on the chart.
It is important to remember the two uses of an R control
(combined with the measurement variation) and the
being able to discover how much of the original variability was due to product.
Using the variables for an x chart fill out the
worksheet
Chief way to control attributes (good or bad) Used for sampling attributes, classified as
defective or non-defective
Percent defective – p chart Number defective – c chart
Upper Limits Control = p + zợp Lower Limits Control = p - zợp P = mean fraction defective in samples Z = number of standard deviations Ợp = standard deviation of the sampling
distribution
Variance equal to its mean Basis for c charts C is number of defects per unit Standard deviation is equal to the square root of
c
Control limits = c +- 3 (square root of c)
Middle line is the target for quality
Managers must select points in process that need
SPC (Statistical Process Control)
Must decide whether variable or attribute charts
are appropriate
Company must set clear SPC policies for
employees to follow
Values must fall within upper and lower control
limits
Typically +- 3 standard deviations from process
mean
Process Capability Ratio = (upper specs – lower
specs)/ 6Ợ
Form of testing Take random samples from batches Measure finished products against predetermined
standards
More economical than 100% inspection Attribute inspection is more common than
variable
Describes how well plan finds good and bad
batches
Want to avoid having good batch rejected
(producers risk)
Consumer wants to avoid bad batch (consumer
risk
Acceptable quality level is lowest level of quality
acceptable
Acceptable level/number of items = AQL
Maximum value corresponds to highest average
percent defective
Acceptance sampling helps increase quality of
batches by reducing outgoing percent defective