The Importance of Viewing Data Over Time in Assessing Baseline Performance - Pt 1

In a previous post, we discussed the importance of establishing a clear understanding of baseline-level performance before taking action on a process. By establishing a clear baseline, organizations can identify variations, monitor process stability, and implement data-driven improvements. Two essential tools for this purpose are Run Charts and Individual Moving Range (I-MR) Charts. These charts are invaluable for visualizing process performance over time, identifying trends, and distinguishing between common cause and special cause variations. This article will dive into the technical aspects of creating a Run Chart, providing a step-by-step guide and best practices. We will continue this theme in Part Two by reviewing Individual Moving Range Charts.

What is a Run Chart??

A Run Chart is a simple graphical tool that displays data points in a time-ordered sequence. It is used to track the performance of a process over time and identify trends, shifts, or cycles. Unlike more complex control charts, Run Charts do not require extensive statistical knowledge to create or interpret, making them accessible to a wide audience.

?Steps to Create a Run Chart

1. Collect Data: The first step in creating a Run Chart is to collect data from the process you wish to monitor. This data should be collected at consistent time intervals to ensure comparability.
2. Plot the Data Points: On a graph, plot the data points in chronological order. The x-axis represents time, while the y-axis represents the value of the process measure you’re tracking.
3. Draw a Median Line: Calculate the median of the data points and draw a horizontal line across the chart at this value. The median line serves as a reference point for analyzing the behavior of the data over time.

Analyze the Patterns:

Once the chart is complete, analyze the data for patterns. Specifically, look for:

• Trends: A sustained increase or decrease in the data points.
• Shifts: A sudden change in the level of data points, indicating a potential change in the process.
• Cycles: Regular patterns that repeat over time may indicate seasonal effects or other cyclical influences.

Rules for Interpreting a Run Chart

Run Charts are powerful tools for detecting patterns that might not be immediately obvious. To interpret a Run Chart effectively, consider the following rules:

• Shift Rule: If you observe six or more consecutive data points all above or below the median, this indicates a shift in the process, which may signify that the process has fundamentally changed.

• Trend Rule: A trend is indicated by five or more consecutive data points all increasing or decreasing. This could suggest a drift in the process that needs investigation.

• Run Rule: The number of runs (a run is a sequence of data points on one side of the median) should be analyzed. Too many or too few runs may indicate non-random variation in the process. This rule is a bit more complicated and is based on a table that depicts a lower and upper limit to the number of runs that should be expected based on a given number of data points.

• Astronomical Points: Any single data point that is significantly higher or lower than the rest should be investigated as it might indicate a special cause of variation.

In Part Two, we will discuss the Individual Moving Range (I-MR) Chart, a slightly more sophisticated way of analyzing data over time.

?

Would you like to learn more about using tools like the one discussed here to improve your products and services? Click the link below to register for one of our upcoming Improvement and Innovation courses or to schedule a chat with one of our representatives.

https://pii.textiles.ncsu.edu/training-certification-levels-overview/