How to Interpret and Use a Control Chart
>>
>>
Manufacturing Explained
>>
How to Interpret and Use a Control Chart

How to Interpret and Use a Control Chart

As a manufacturer, you need to know if your processes are stable and in control or unstable and out of control. This is imperative in measuring, tracking, and maintaining quality.

New Feature Alert: We recognize this importance; control charts are now a feature within Nulogy Smart Factory.

In this blog, we’ll talk about what a control chart actually is, how it’s used, and finally, walk you through how you’d use a control chart within Nulogy Smart Factory.

What is a Control Chart?

Control chart: A control chart is one of the primary techniques of statistical process control (SPC). The control chart is a graphical display of quality characteristics that have been measured or computed from a sample versus the sample number or time. The control chart was invented by Walter Shewhart at Bell labs in 1920.

Statistical Process Control: Statistical Process Control (SPC) is a statistical method to measure, monitor, and control a process. In other words, SPC is a method of quality control that employs statistical methods to measure, monitor, and control a process.

The concept of Statistical Process Control (SPC) is a primary tool, prescribed by the methodologies of Six Sigma, to improve quality by reducing process variation.

A control chart plots data over time to identify fluctuations in a process.

The keyword here is process control. Let’s say there is a process with a metric you want to track and ensure stays stable. Maybe, it’s cycle time, for example. A control chart will monitor cycle time for any fluctuations.

Another example is temperature. Maybe it’s a refrigerator and you need that temperature to stay within a particular range of control. You would use a statistical process control chart to monitor these metrics, and more.

New Feature: Control Charts in Nulogy Smart Factory

Within Nulogy Smart Factory, manufacturers now have the ability to use control charts to understand process changes over time. Is a particular process out of control or not?

These graphs are typically used in accordance with preventive maintenance, providing a critical measurement that will tell if the product is going to be of good quality or not. If a measurement is not within pre-set boundaries, there is a potential quality issue.

There are many types of control charts, each with a specific purpose. Knowing which chart will be most beneficial is going to be dependent on what is being produced at the plant. 

A few weeks ago, we added this capability to the Nulogy Smart Factory platform. We now support 3 different control charts, with greater detail provided in the next section.

What We Support

We support 3 types of control charts: X-Bar, R, and S charts. Typically, the control charts are viewed in pairs, either an X-Bar with an R chart or an X-Bar with an S chart.

control chart x-bar chart
X-bar chart: Defined by SixSigmaStudyGuide.com, an X-bar chart is “the mean or average change in process change in the process over time from subgroup values. The control limits on the X-bar bring the sample’s mean and center into consideration.”
control chart r bar chart
R chart: This is best used for smaller subgroup sizes, typically 2-10.
control chart s-bar chart
S chart: This is best used for larger subgroup sizes, typically great than or equal to 11.

These types of charts can be used in a lot of places. Basically, anything you want to control and make sure remains consistent can be monitored by a control chart.

So, let’s dive into greater detail. In an X bar chart, X is a sample value like cycle time or temperature. Each one of the dots is an average of a certain number of samples in a subgroup. This average of X is called X Bar.

The gray center line is the Mean (or average) of those averages, also called the X Double-Bar. The red lines above and below are the Upper Control Limit (UCL) and Lower Control Limit (LCL). The UCL and LCL are calculated with a standard formula. Ideally, the subgroup averages, that is, the dots, should remain within those limits. Any X Bar value that falls above the upper limit or below the lower limit could be considered “out of control”.

The R Chart is used to visualize how widely the process varies over time. Each dot on the R Chart represents the range of values in each subgroup. This range of X is called R. And, like the X Bar chart, the Mean, UCL, and LCL represent the average of the ranges (also called R Bar) and the upper and lower control limits.

The S Chart is similar to an R Chart in that it measures the variation of the process. But instead of using a range of values, each dot represents the standard deviation of the values in each subgroup. This standard deviation of X is called S. And like the X Bar and R Chart (you guessed it!), the Mean, UCL, and LCL represent the average of the standard deviations (also called S Bar) and the upper and lower control limits.

The purpose is to look at all of these values together to provide a complete picture of what’s happening. It’s an indication of how stable the processes are.

In Nulogy Smart Factory, the data is measured at a continuous pace, sampling and measuring data at 5-minute intervals. This helps manufacturers identify variations over time.

Related reading

Related Posts

Manufacturing Downtime

Built for CNC or built for your whole line? How to evaluate machine-monitoring software fit

Understand how to choose the right machine monitoring software fit for your production needs to enhance efficiency and visibility.
Nulogy Smart Factory dashboard TV monitor

Visual scoreboards vs. AI-assisted production monitoring: what’s actually different

Explore visual scoreboards vs AI production monitoring and discover which approach better enhances manufacturing efficiency.
Manufacturing line tablet collaboration managers

The hidden cost of monitoring software that outgrows your plant, not the other way around

Explore production monitoring software that scales efficiently for multiple lines and sites, reducing integration costs.