How do you monitor your data over time with control charts?

1 What are control charts?

Control charts are graphical displays of data that show the variation of a process over time, consisting of a central line representing the average or target value and upper and lower control limits defining the range of natural variation. As data points are collected, they are plotted on the chart and any patterns or trends are observed. Control charts can help you determine if the process is stable and predictable, meeting customer specifications or requirements, capable of producing consistent and high-quality outputs, or if there are any signals of special causes of variation that need to be investigated and eliminated.

2 How to choose a control chart?

When selecting a control chart for your data, you need to consider the type of data (continuous or discrete), the sample size and frequency of data collection (large or small, constant or variable), and the objective of the analysis (to monitor the central tendency, the variation, or both). The most common control charts are variable control charts, used for continuous data that can be measured on a scale, such as length, weight, or temperature. Examples include X-bar and R charts which monitor the mean and range of subgroups of data, and I-MR charts which monitor individual values and moving ranges. Attribute control charts are used for discrete data that can be counted or classified into categories, such as defects, errors, or pass/fail. Examples include P charts which monitor the proportion of defective items in a sample, and C charts which monitor the number of defects per unit.

3 How to construct a control chart?

Constructing a control chart requires collecting data from the process you want to monitor, preferably in a random and representative way. This data should cover at least 20 to 25 subgroups or individual values. Then, you need to calculate the central line, usually the mean or median of the data, and the upper and lower control limits, which are usually based on the standard deviation or other measures of variation. Plot the data points on the chart along with the central line and control limits. You can use a software tool or spreadsheet for this purpose and update it as new data becomes available. Lastly, analyze the chart for any signs of instability or abnormal variation, such as points outside the control limits, runs of points on one side of the central line, or trends or cycles in the data. These indicate special causes of variation that need to be investigated and corrected.

4 How to use a control chart?

A control chart is not only a tool for displaying data, but also a tool for decision making and improvement. You can use it to compare the process performance to customer specifications or requirements and determine if the process is capable of meeting them. This is done by adding specification limits to the chart, which represent the acceptable range of variation for the customer. Additionally, you can use a control chart to identify the sources of variation in the process and distinguish between common and special causes. Moreover, you can monitor the process over time, and evaluate the impact of any changes or improvements made to the process. Control charts can be used to track changes in the central line, control limits, or data patterns and see if they result in a better or worse performance; as well as verify if the process remains stable and in control after improvements are implemented.

5 What are the benefits of control charts?

Control charts are a valuable tool for any Lean Six Sigma project, as they can help to improve the quality and consistency of products or services, increase customer satisfaction and loyalty, reduce costs and waste, enhance process efficiency and effectiveness, and foster a culture of continuous improvement. These charts can do so by reducing variation and defects in the process, meeting or exceeding customer expectations and requirements, avoiding rework, scrap, or inspection in the process, optimizing process parameters and inputs, and encouraging data-driven decisions and problem-solving.

6 Here’s what else to consider

This is a space to share examples, stories, or insights that don’t fit into any of the previous sections. What else would you like to add?