What Is A Control Chart? (With Types And Steps To Plot One)

By Indeed Editorial Team

Published 19 July 2022

The Indeed Editorial Team comprises a diverse and talented team of writers, researchers and subject matter experts equipped with Indeed's data and insights to deliver useful tips to help guide your career journey.

A control chart is a statistical tool, which a business or a manufacturing unit uses to identify variations in processes. With the help of these charts, an enterprise evaluates its actions and makes decisions accordingly to improve a process or a system. Learning about control charts can help you interpret them better, along with monitoring the behaviour of processes and achieving and maintaining process stability. In this article, we answer the question, 'What is a control chart?', discuss why teams use it, list its different types and outline the steps to plot one.

What Is A Control Chart?

The answer to the question, 'What is a control chart?' is that it is a quality analysis tool that organisations use to determine if processes are operating in statistical control. Control charts help enterprises to visualise the variations in the process data and its components. It shows whether the process is in a stable state or out of control. With this tool, enterprises distinguish between common causes and special causes. Here is what these two types of causes of variation in a measure mean:

Common causes

Referred to as random or unassignable variation, common causes exist naturally in a system and have the potential to affect everyone working in the system, along with affecting the results of the system. Common causes occur because of regular or natural causes. These causes affect all outcomes of a process, resulting in a stable dispersal that is predictable.

For instance, you take an hour to reach the office every morning. The time you take to reach the office is not exactly 60 minutes. Due to variations, you may take 50 minutes or more than an hour, given there are variations in common causes, like the number of traffic signals or the volume of the traffic.

Special causes

Special causes are not inherent in the system and do not affect people in the system all the time. These causes result from certain circumstances. Referred to as non-random or assignable variation, these causes result in an unstable process that is not predictable.

For instance, usually, you take 60 minutes to reach home from your office. On one particular day, you take two hours to cover the same stretch. This may be because of a road that authorities have closed or there is an accident in the area that is causing a heavy traffic jam.

Related: What Is Quality Control? A Complete Guide

Uses Of Control Charts

Control charts allow teams to evaluate the behaviour of a process to determine its stability. Similar to run charts, control charts show data in a time sequence, but are more efficient than run charts to assess and achieve process quality. Different teams of an organisation may use control charts for various purposes:

  • Monitoring processes or events over time to identify when the developments are higher or lower than expectations

  • Keeping processes stable and eliminating special causes

  • Reducing the number of undesirable events or increasing the count of desirable events

  • Monitoring to see if there is a decline in the variability of processes to determine improvement in the quality

  • Differentiating between the variation in special causes and common causes

  • Making decisions regarding the specifications of the products and process of the production

  • Identifying and addressing root causes of variation in processes

  • Measuring if the process improvements are aligning with desired goals

Related: 10 Quality Tools To Use (With Definition, Benefits And Tips)

Types Of Control Charts

Attributes data and variables data are two categories of control charts. Variables data control charts include X-bar and R charts for sample averages and ranges, X-bar and S charts for sample means and standard deviations, Md and R charts for sample medians and ranges and X-bar charts for individual measures. Attributes data control charts include c charts to count non-conformities, p charts to determine the proportion of units non-conforming, np charts to find out the number of units non-conforming and u charts to count non-conformities per unit. Here are some commonly used types of control charts for data analysis:

X-bar and R control chart

An X-bar and R chart helps organisations to monitor the mean and variation of a process. This is based on the samples teams take from a process at a specific time. The measurement of these samples from a sub-group helps the teams to evaluate the mean and standard deviation of a process. The team members then use the mean and standard deviation to create control limits for the range of each sub-group, along with the mean.

X-bar and S chart

An X-bar and S chart utilises the standard deviation of sub-groups instead of the range for evaluating the standard deviation. Teams generally use these charts when the sample size of the sub-group is relatively large. For instance, if the sample size is larger than ten. It helps teams to monitor the process average and standard deviation efficiently.

P and np control charts

P and np control charts allow teams to evaluate variations in a pass or a fail data type. These potential outcomes show whether a process is defective or not. Teams use p charts to assess the proportion of non-conforming units based on samples. When teams establish the control limits, they use these limits to evaluate the non-conforming proportion of the process.

C and u control charts

Both C and U charts help teams to analyse the variation in the number of defects in a sub-group. To evaluate the average number of non-conforming units per sample, teams can use an initial series of samples. In this phase, if a process is not in control, the teams then determine the problems in a process to remove the defective units.

Related: What Is Process Control? (Definition, Benefits And Examples)

How To Plot A Control Chart

Below are some steps to help you plot a control chart:

1. Choose a measurement method

Variables and attributes are two types of data that control charts use. With variables, teams have access to detailed information, which they use to solve problems. With attributes data, teams get data classifications, such as conforming and non-conforming. This type of data is good for making yes or no decisions.

2. Validate the data

Before using the data in your control charts, ensure that the data you have collected is correct. Wrong data affects the measurement, which results in flawed decision-making. If you plan to create effective control charts, validating the data periodically proves to be helpful.

3. Calculate control limits

After you have collected and validated the data, calculate the average of the data. Insert a control line, along with upper and lower control limits, once you have calculated these. Consider using upper and lower control limits in different colours to make a coherent chart.

To calculate control limits, subtract the average number from the number you have recorded. Then, square the result and take out an average of all the squared results. Find the square root of this average to determine the standard deviation. Decide the number of standard deviations for your controlled process. A process that is in control has data which is within three standard deviations from a mean.

4. Identify root causes

If a process is out of control, it means that points are outside of your control limits and there may be some special causes affecting the process. Determine these causes and analyse the variations. This analysis helps you to find the solutions for the variations and enhance the quality and efficiency of the processes.

Related: Problem-Solving Skills: Definitions And Examples

Different Calculations Of Control Charts

There are many varieties of control charts with different calculations, but all these charts involve:

  • Calculation of an average for a process, which is generally the mean

  • Calculation of a measure of variation in a process, which is generally the standard deviation

  • Calculation of the standard error, which teams may perform using standard deviation and sample size

  • Calculation of upper control limit (UCL) and lower control limit (LCL), which is generally set at a couple of errors above or below the mean

  • Data that a team collects and plots over time to determine process improvements or downturns and variability reduction

Related: Core Strengths Needed To Succeed In The New World Of Work

Ways To Interpret Control Charts

Finding out if a system is stable or unstable and making decisions based on this knowledge are two methods to interpret control charts:

Determine if the system is stable or unstable

When a process has only common cause variations, it is a stable process and when a system has both common and special cause variations, it is an unstable process. To advantageously interpret control charts, detect special cause variations and gain insights into factors that are affecting the corrective measures. Conduct monthly team meetings to analyse data. This helps teams to test the relationship between data and measures the team is taking to improve the system. This assessment of control charts enables an organisation to evaluate the effectiveness of processes.

Make decisions

When teams differentiate between the two types of variation, they make excellent decisions. If the system is stable, a team may determine the factors that are resulting in variations common to all the points. This helps teams to develop a system of performance that benefits them. If the system is unstable, a team may better understand the circumstances that resulted in special causes. This assessment helps team members to recognise the changes that are leading to the special causes. The identification of these causes enables them to implement changes and deliver better system performance.

Explore more articles