Variance Analysis analyzes the selected parameters, ranks them by variance, and displays summary metrics for the top 10 parameters. This tool is most helpful early in the exploratory process to find parameters that may be good candidates for additional analysis or to identify parameters with minimal or no variation. This is a good tool to lead to Descriptive Statistics or Time-Series Correlation.
Variance Analysis is a multivariate analysis that you can perform on continuous fields on the Cycle model. You can select multiple machines of a given type.
Variance Analysis computes the variance of all selected parameters over your selected date range. After ranking the parameters by their variance, the tool selects the top 10 parameters. For each parameter, the tool plots a histogram of their values over the date range and a table of summary statistics including Count, Standard Deviation, Minimum, and Maximum.
How to Use a Variance Analysis
The Variance Analysis tool is an exploratory data tool that identifies parameters that have high levels of variance and are therefore good candidates for additional analysis. It is often one of the first steps in performing data analysis.
NOTE: This tool only works on numeric fields, not categorical fields such as Pass/Fail.
This tool is built on a Cycles model so it ignores variability that happens outside the cycle (non-production data), which could interfere with the analysis vs. traditional variance analysis done via spreadsheets.The Cycles model summarizes parameter readings for each cycle, eliminating noise from higher granularity readings that may be too detailed for an exploratory analysis. Raw data can be analyzed if needed using the Raw Data Visualization tools. For more information, see Generating a Raw Data Visualization.
To use a Variance Analysis:
- On the Analysis tab, under Process Variability, click Variance Analysis.
- On the main Variance Analysis screen, select your options on the left. For more details about each option, see Variance Analysis Options.
- Click Update.
Variance Analysis Options
The Variance Analysis options include:
- Model: You can analyze only cycles for a specific asset (that is, machine). You cannot change this option.
- Assets: You can select one or more assets of the same type to analyze.
- Relative Range: You can define a date range of data points to analyze for the assets you have selected. For example, select the last 7 days. This option establishes the boundaries for the near real-time data that will be part of your analysis. Select from relative or absolute timeframes.
- Data Fields: You can select data fields to analyze.
- Carry-forwards: You can undo the forward-fill effect on any fields. Select from Keep All, First, or Last. For more information, see the table in Options Common to All Analysis Tools.
- Update: Click this button to generate your chart.
The Variance Analysis tool displays and ranks the 10 parameters from this asset that exhibit the highest amount of variability.
A histogram for each parameter lets you visualize the overall distribution of its values. A box plot for each parameter lets you visualize the overall distribution, quartiles, and the standard deviation. This helps you determine the top parameters that are good candidates for further investigation.