Quality tools continued. Flashcards
Which guru is associated with the cause and effect diagram?
Ishikawa
Which guru is associated with control charts?
Walter A. Shewart
What are control limits based on?
The process
What is the objective of a control limit?
To determine if the process is in statistical control.
What is a specification limit?
Established by management
What is the objective of a specification limit?
Ensure product meets design or regulatory requirements.
What are control charts used for?
To monitor, control, and improve process performance over time by studying variation and its sources
What are common causes of variation?
- Variation stable and predictable
- Small changes in process factors
- Raw materials
- Normal
What are special (attributable) causes of variation?
- Variation unstable and unpredictable
- New cause that was previously absent
- e.g., wrong ingredient; wrong process setting; untrained operator
Why is it necessary to measure variation?
The ability to measure variation is necessary before it can be controlled
What are the 3 categories of variation?
- Within piece
- Piece to piece
- Time to time
What are 4 sources of variation?
- Equipment
- Material
- Environment
- Operator
What happens if you ascribe variation to a special cause that is actually the result of a common cause?
- Mistake 1
- Over adjustment
What happens if you ascribe variation to a common cause when it is actually the result of a special cause?
- Mistake 2
- Doing nothing
Describe the ‘2 mistakes’ in controlling variation.
- Mistake 1 is to ascribe variation to a special cause when it is the result of a common cause (e.g.. Over adjustment)
- Mistake 2 is to ascribe variation to a common cause when it is the result of a special cause. (e.g.. Doing nothing)
What three basic components do all control charts have?
- a centerline, usually the mathematical average of all the samples plotted.
- upper and lower statistical control limits that define the constraints of common cause variations.
- performance data plotted over time.
Describe variable control charts.
- Used in pairs
- X-bar and R-bar chart
- Control measurable variables
- e.g., pH, %solids, grams etc.
Describe attributable control charts.
- Based on a characteristic of a product, process, or population that can be counted or tallied, but is not described in incremental numbers.
- Characteristic that is:
- Defective or non-defective
- Satisfactory or unsatisfactory
- Heavy or light
When would you use an x-bar control chart? [4]
- To indicate when observed variations in quality are greater than could be left to chance
- A way of visualizing the variations that occur in the central tendency and dispersion of a set of observations
- Shows whether the process is in a stable state
- It is a graphical record of the quality of a characteristic
What is the central line in an X-bar control chart?
- Average of the plotted points , i.e. the average of the averages or “x-double bar”
- Standard or reference value based on prior data or specifications
What are upper and lower control limits, and what are they used for?
- They are a function of the subgroup averages, and are usually established at +/- 3 SD from central line.
- They aid in judgement of the significance of the variation
- They are used to evaluate the variations in quality from subgroup to subgroup
Do not confuse with specification limits!!!
Control charts are used to identify and separate chance random variation from variation due to attributable cause. Special causes (i.e., unnatural variation that is the result of attributable causes) must be identifed and eliminated.
How is this done?
- Notice where the data exceeds either the UCL or the LCL.
What are the 6 steps in creating a control chart?
- Select the quality characteristic
- Choose the subgroup
- Collect the data
- Determine the trial central line and control limits
- Establish the revised central line and control limits
- Achieve the objective
What three issues are important when sampling for control chart purposes?
- Subgroup (sample) size considerations
- Sampling frequency
- Process of collecting samples
What is done for each subgroup sample collected for a control chart?
- Subgroup average
- Subgroup range
- Plot over time
What is characteristic of a stable process?
- Most of the points are near the centerline
- Points occur at random with decreasing frequency as they approach the control limits.
When is the process in control?
When the outcome is predictable within probabilistic limits.
When is a process out of control?
The process is unpredictable.
How is a histogram made in Excel?
- To create a histogram in Excel, you provide two types of data — the data that you want to analyze, and the bin numbers that represent the intervals by which you want to measure the frequency.
How is the centering of a histogram interpreted?
How is the variation of a histogram interpreted?
What action would you take if a histogram is centered and well within customer limits?
Maintain present state
What action would you take if a histogram demonstrates no margin for error?
Reduce variation
What action would you take if histogram shows a process is running low and there are defective products/services.
Bring average closer to target
What action would you take if a histogram shows the process is too variable and there are defective products/services?
Reduce variation.
What action would you take if a histogram shows a process is off center, too variable, and there are defective products/services?
Center better and reduce variation
What is the difference between a bar chart and a histogram?
Bar chart: used to compare variables; plot of categorical data
Histogram: plot of quantitative data with ranges of the data grouped into bins or intervals; shows the distribution of variables
What is the 80/20 rule?
- Focus on key problems
- The Pareto Principle of time versus result
- Focus on the ‘vital few’ and not the ‘trivial many’
How do you determine correlations with a scatter plot?
Which variable goes on the y-axis?
Dependent
