Statistical Tools
The mission of Six Sigma at Weston Aerospace is to ensure that Six Sigma ‘thinking’ is applied wherever data is used.
In recognizing the potential rewards of applying Six Sigma at Weston, we are following in the footsteps of many world class companies who have realized significant returns from the methodology.
Shown above: Example results of a Gage Repeatability & Reproducibility study. Here the measurement system variation is compared to the process tolerance, allowing the suitability of the measurement system to be evaluated.
The Six Sigma Toolkit
The Six Sigma toolkit is based around five stages of a project:
1.Define – Who and what is involved? What is expected from the project?
2.Measure – Evaluate the current process and measuring system.
3.Analyse – Find out what factors are important in the process.
4.Improve – Find out how to run the process to deliver required performance.
5.Control – Maintain improved process.
At Weston, specific projects are used not only to implement Six Sigma methodologies and improve performance, but also as the means to educate the project members in the tools and techniques used.
The data-driven method uses tools such as those shown below:
Thought maps are used to track project progress and to document what is known about a process. Information that is required is also detailed, as well as the method by which it will be gathered.
Process maps, essentially a flow diagram of the process as it happens, are used to evaluate all sources of variation. Example Process Map:
Used to establish suitability of measurement systems. Example image from Minitab computer program:
Used to assess capability of a process against specification limits. Example image from Minitab computer program:
Used to establish if effect seen is statistically significant.
Graphical representations for use of Hypothesis Testing:
F-Test – used to establish if the spread of the data has statistically changed
T-Test – used to establish if the average of the data has statistically changed
Used to establish if major sources of variation are within or between groups of data. In this example image from the Minitab computer program, the majority of points fall outside the control limits on the X-bar chart, so the majority of variation is between subgroups:
Used to establish the important factors in a process. The following example image from the Minitab computer program, shows the effects of changing 3 factors – A, B & C – on the measured output of a process:
Cube plot – used to graphically evaluate the main effects and interactions between 3 factors in a designed experiment.
Used to establish and monitor normal and special causes of variation in a process.
Example image from Minitab
computer program:
Weston Aerospace / 124 Victoria Road / Farnborough / Hampshire / GU14 7PW / United
Kingdom
