- Back -
Production Data to Results in 4 Simple Steps:
This example explains steps involved in recognising patterns from production data on alloy chemistry.
| 1 |
Data Collection
Chemical composition data was recorded on 87 heats for a high performance nickel based alloy and % rejections due to shrinkage defects were recorded for castings produced in each heat.
View Production Data (Excel format) |
 |
| 2 |
Convert Data in MetaCause Format
Once the production data is downloaded from your database, scatter diagrams are plotted for each factor and response to decide threshold values that are used in MetaCause analysis.
On its own, this data has little meaning and must be analysed to extract the information
View MetaCause Data (Excel format) |
 |
| 3 |
MetaCause Analysis
The MetaCause turns the raw data into clear and accurate information.
|
 |
| 4 |
Results
The results are clear, accurate and easy to interpret
|
 |
The Results
Main Effects: For each response, factor settings are classified as either Optimal Settings, Settings to Avoid or Settings with No Effect.
Factor settings with importance weighting value around 50 and more have quantifiable influence on the response. |
Report Output |
Interactions: Optimal Interactions and Interactions to Avoid are identified for every response.
A factor setting may not have significant influence on the response on its own however, it may interact with another setting to produce a quantifiable influence. |
Report Output |
Analysing data using MetaCause gives you clear information on process improvement opportunities almost instantly. The simple to use software applies very advanced techniques enabling you to identify factors that need further monitoring, and react quickly.
MetaCause can improve your analysis and save you money.
- Back - |
