Our World Statistics Day conversations have been a great reminder of how much statistics can inform our lives. Do you have an example of how statistics has made a difference in your life? Share your story with the Community!
Choose Language Hide Translation Bar

Effective Presentation Methods for 3-D Data Using JMP®

Author:   Michael Thompson, Group Leader - Advanced Gauging Systems, The Timken Company



Presenting three-dimensional data in useful and effective ways offers challenges regardless of the visualization platform. This paper will explore several methods (and their combination) that can be used within JMP to simply and effectively present three-dimensional data sets.


  1. Use of 3D surface, mesh, and scatterplots. These projection-based tools commonly used for 3D visualizations and analyses within JMP are first explored using a complex undulating surface defined precisely by a mathematical function.
  2. Use of contour plots to flatten a three-dimensional data set to two dimensions. The same function-based data used to present surface, mesh, and scatterplots is then plotted as a contour plot. A contour plot slices the data into planar groupings and produces outlines/regions that define the boundaries/contours produced by those slices.
  3. Use of a color scale to represent the third dimension. To demonstrate this technique, data representing wear scar development on a calibrated precision gauge block is presented with position as one dimension, calibration cycle number (time history) as the second dimension, and wear scar depth as the third (color) dimension.
  4. Use of grouping to show progression along a third dimension. For this demonstration, the gauge block wear scar data is presented again using grouping over time in order to show how the gauge block starts with no wear and then is progressively worn through its repeated use. The grouping, in this case, acts as a form of averaging over defined intervals of time.
  5. Use of marker transparency and size to produce a pseudo-density as the third dimension. To demonstrate this technique, multiple scan data sets acquired with a tactile profilometer are overlaid in order to show the repeatability of the measurement system. When the multiple data sets overlay closely, markers will completely or partially hide other markers causing knowledge of data density (and consequently scan agreement) to be lost via marker saturation. In cases like this, use of marker transparency and size can help to highlight the agreement and deemphasize outlier and/or errant readings. The plot produced is effectively a density plot where marker transparency and size are used in combination as a mechanism to generate a density dimension.
  6. Use of time (animation) to represent a third dimension. For the final example, two-dimensional scan data sets from repeated CMM (Coordinate Measuring Machine) calibrations are concatenated into a single large JMP table. Each scan data set shows the condition of the probe tip (as measured in a single planar slice) using a reference sphere. Over time, wear scars develop on the probe tip due to its use at preferred contact angles. The development of these wear scars is animated within JMP through the use of the Local Data Filter.


Several of these methods (both individually and in combination) are easy to perform within the JMP Graph Builder. Due to its versatility and intuitive user interface, the JMP Graph Builder should be one of the first platforms the JMP user launches for data visualization, analysis, and exploration.


Refer to attachments for full document and associated JMP tables.