Statistical Quality Control

Front 1

TWO FOCUSES FOR STATISTICAL QUALITY CONTROL

(assessing the quality of an ongoing process vs. assessing the quality of an existing batch

Back 1

Assessing The Quality Of An Ongoing Process - This facet is called statistical process control, and relies on the use of process control charts to determine whether the ongoing process is performing acceptably.

Assessing The Quality Of An Existing Batch Or Lot Of Items - This facet is called acceptance sampling, and relies on the use of acceptance sampling plans to determine whether the existing lot is of acceptable quality.

Front 2

common (random) causes of variation in a process:

Back 2

Every process has some degree of inherent, natural (or random) variation. This type of variation is unavoidable, and cannot be removed from the process.

Front 3

assignable (special) causes of variation in a process:

Back 3

Sometimes process variation can be directly attributed to a specific or assignable cause (often called a special cause). This type of variation can be eliminated from the process if the special cause can be found and corrected.

Front 4

What do all Control charts have in common?

Back 4

they record the results of successive samples taken over time. Each has a center line that reflects the process mean for the aspect of the product or service that is being monitored. Each also has an upper control limit (UCL) and a lower control limit (LCL), which define the range of variation that is presumed to be random variation.

Front 5

the place where all control charts differ.

Back 5

The control charts differ only in the aspect of the product or service produced by the process that is being monitored. This information is reflected on the vertical axis of the chart.

Front 6

attributes measurement

Back 6

An attribute can be evaluated with a discrete response such as good or bad, acceptable or not, yes or no, works or doesn’t work, and sometimes evaluated by a simple count. An attribute evaluation is sometimes referred to as a qualitative classification.

Front 7

Variable measurement

Back 7

A variable measure is a characteristic of a product or service that is measured on a continuous scale, such as length, weight, temperature, or time. A variable evaluation is sometimes referred to as a quantitative classification.

Front 8

P-chart

Back 8

It measures the proportion that is defective in the samples.

Front 9

C-chart

Back 9

It records a simple count of the number of occurrences of some event in the samples.

Front 10

Mean chart

Back 10

(also referred to as an x-bar chart) It measures and records the value for some continuous variable associated with the product or service as the process is sampled.

Front 11

Range chart

Back 11

also referred to as an R chart). It measures and records the range of values for the continuous variable being monitored as the process is sampled.

Front 12

What is the screening process?

Back 12

100% inspection. in practice, screening still might overlook some defective items, for no inspection is perfect

Front 13

Reasons to Avoid Screening

Back 13

1. Cost of inspection is high relative to the cost of a defective item.

2. Inspection is very time consuming and labor intensive (which indirectly translates into costs).

3. Nature of the test may require the destruction of the item to determine whether it was acceptable.

Front 14

The parameters of a single sampling plan

Back 14

It is used to make a judgment about the quality of an existing batch, or lot, of items.

if x <= c then accept

Front 15

Producer Risk

Back 15

risk (Type I errors) often result in internal failure costs, especially when rejected lots (which were in fact good) are subjected to additional scrutiny to try to find problems (that are not even present), or when the producer simply scraps the rejected lots and starts over from scratch.

Front 16

Consumer Risk

Back 16

(Type II errors) tend to result in external failure costs, for the result of such an error is nonconforming output ending up in the hands of the consumer.

Front 17

OCC

Back 17

An operating characteristic curve is a device that shows a sampling plan’s ability to discriminate between good and bad lots.

Front 18

diffuse

Back 18

v. to spread out, wordy