Statistical Process Control (SPC) is a quality control method used to monitor and control a process through statistical analysis. SPC helps identify and correct problems in a process before they result in defects or non-conformities. By analyzing data and establishing control limits, SPC determines whether a process is stable and predictable, or unstable and unpredictable, allowing for corrective action to be taken if necessary. SPC is widely used in manufacturing industries but can also be applied to other processes, such as service industries or healthcare. In this article, we will explore the key concepts of SPC and its applications, as well as the benefits and limitations of the method.

Streamlining Statistical Process Control (SPC) with InspectionQuest®

A Brief History of Statistical Process Control

Statistical Process Control (SPC) is a powerful quality control technique used to monitor and control a manufacturing process. It helps manufacturers identify and eliminate defects, reduce variation, and improve overall product quality. SPC was introduced to the world by Walter Shewhart and W. Edwards Deming in the 1920s and 1930s, and it has been widely used in Japan since the 1950s to improve quality in various industries.

Walter Shewhart was a physicist and statistician who worked for Bell Telephone Laboratories in the 1920s. He developed the idea of statistical control charts, which he used to monitor the production of telephone parts. The charts helped him identify when a process was operating within its normal limits, and when it was producing defective parts. Shewhart’s work formed the basis for modern SPC.

W. Edwards Deming was an American statistician who worked in Japan in the 1950s to help improve their manufacturing processes. Deming taught the Japanese the principles of SPC, which they applied to their industries with great success. Japan’s use of SPC was a key factor in its economic recovery after World War II.

How Complex, High-Volume Industries Apply SPC Today

SPC is now used in many industries, but it is particularly important in the automotive and aerospace industries. These industries require extremely high levels of quality control, and SPC helps them achieve this. In the automotive industry, SPC is used to monitor and control tight-tolerance characteristics in the production of cars, trucks, and other vehicles. In aerospace, it is used to monitor and control critical characteristics in the production of aircraft and spacecraft.

The Automotive Industry Action Group (AIAG) has developed guidelines for Statistical Process Control (SPC) specifically for the automotive industry. These guidelines provide a standardized approach to SPC that helps manufacturers improve quality, reduce defects, and increase productivity. The AIAG SPC guidelines cover topics such as data collection, control chart selection, process capability analysis, and control plan development. They also provide detailed instructions on how to use SPC software to automate the data collection and analysis process. By following these guidelines, automotive manufacturers can ensure that their SPC processes are effective and consistent across their organization.

Similarly, the International Aerospace Quality Group (IAQG) has developed the AS9145 standard for APQP and PPAP, which provides guidance on the use of Statistical Process Control (SPC) and initial capability studies in aerospace as part of the APQP process. This standard is designed to help organizations in the aerospace industry achieve consistency and quality in their processes through the use of advanced quality techniques such as SPC and capability analysis. It covers topics such as control charts, process capability analysis, and statistical analysis. By using the AS9145 standard, organizations can align their SPC processes with industry best practices and the requirements of their customers, ensuring the quality and consistency of their products and services.


Common Statistical Tools Used to Monitor Process Variation

The basic idea behind SPC is to monitor a manufacturing process using statistical tools. The process data is collected and analyzed using control charts, histograms, and other statistical tools. These tools help identify when a process is operating within its normal limits, and when it is producing defective products. By identifying defects early, manufacturers can take corrective action before large quantities of defective products are produced.  Of the tools used, control charts, specifically the X-bar and R chart, are the most common.

There are several types of control charts that can be used in Statistical Process Control (SPC), each designed to monitor a specific aspect of a manufacturing process. The most common types of control charts include:

  1. X-bar and R chart: This control chart is used to monitor the mean and range of a process. The X-bar chart shows the average value of a process over time, while the R chart shows the range of values within each subgroup. It is used when the data is continuous and normally distributed.
  2. Individual and Moving Range (I-MR) chart: This control chart is used to monitor the variability of a process. The individual chart shows the range of values within each subgroup, while the moving range chart shows the range between consecutive subgroups. It is used when the data is continuous and the sample size is small.
  3. p-chart: This control chart is used to monitor the proportion of defective items in a sample. It is used when the data is categorical and the sample size is large.
  4. c-chart: This control chart is used to monitor the number of defects per unit in a sample. It is used when the data is categorical and the sample size is small.
  5. u-chart: This control chart is used to monitor the number of defects per unit of time or volume. It is used when the data is continuous and the sample size is variable.

The choice of control chart depends on the type of data being collected, the sample size, and the purpose of the analysis. For example, the X-bar and R chart is commonly used for monitoring the mean and range of a process, while the p-chart is used for monitoring the proportion of defective items. It is important to select the appropriate control chart based on these factors to ensure that the analysis is accurate and effective.

Understanding Process Variation Seen in Control Charts

In Statistical Process Control (SPC), there are two types of variation that can affect a manufacturing process: common cause variation and special cause variation.

Common cause variation is the natural variation that is inherent in a process. It is also referred to as random variation or systemic variation. This type of variation is caused by factors that are common to the process, such as variations in raw materials, temperature, humidity, machine wear and tear, and human error. Common cause variation is expected and can be managed through process control measures such as process standardization and control charting.

Special cause variation, on the other hand, is caused by factors that are not part of the normal variation in a process. It is also referred to as assignable variation or non-random variation. This type of variation is caused by factors such as equipment malfunctions, operator error, or changes in the raw materials. Special cause variation is unexpected and can lead to defects or other issues in the manufacturing process. It requires investigation and corrective action to identify and eliminate the root cause.

In SPC, the goal is to reduce the variation in the process to improve quality and increase efficiency. Common cause variation can be managed through process control measures, while special cause variation requires root cause analysis and corrective action. By monitoring and controlling both types of variation, manufacturers can ensure that their processes are running at peak efficiency and producing high-quality products.

Limitations and Challenges Associated with Implementing SPC

Although Statistical Process Control (SPC) is a powerful tool for monitoring and improving processes, it also has some limitations. Here are some of the limitations of SPC:

  1. Requires data and analysis expertise: SPC relies heavily on statistical analysis and requires a significant amount of data to be collected and analyzed. This can be a challenge for organizations that lack the necessary data collection and analysis expertise.
  2. Not suitable for small data sets: SPC is best suited for large data sets and may not be appropriate for small data sets. With small data sets, it may be difficult to identify patterns or trends and establish reliable control limits.
  3. Cannot account for all sources of variation: SPC is designed to detect and correct variation that is inherent in a process. However, there may be sources of variation that are not accounted for by SPC, such as changes in the environment or human error.
  4. Limited by the quality of data: SPC relies on the quality of the data collected. If the data is inaccurate or incomplete, SPC may not be effective in detecting process problems.
  5. Limited by the assumptions made: SPC is based on certain assumptions about the process being monitored. If these assumptions are not met, the results may be unreliable and misleading.
  6. Not a substitute for process improvement: SPC is a tool for monitoring and controlling a process, but it is not a substitute for process improvement efforts. Organizations still need to make efforts to improve their processes to achieve better quality and productivity.

Overall, SPC can be a valuable tool for process improvement, but it is important to be aware of its limitations and to use it appropriately.

The Use of Software for Statistical Process Control

The use of software can help to address many of the challenges associated with Statistical Process Control (SPC). Here are some examples of how software can help:

  1. Data collection and analysis: SPC software can automate the data collection and analysis process, reducing the need for manual data entry and analysis. This can save time and reduce the likelihood of errors.
  2. Large data sets: SPC software can handle large data sets more efficiently than manual methods, making it easier to identify patterns and trends and establish reliable control limits.
  3. Variations and assumptions: SPC software can help to identify variations that may not be apparent through manual analysis and can help to test the assumptions that underlie the SPC process.
  4. Quality of data: SPC software can provide data validation checks and error detection, ensuring that the data collected is accurate and complete.
  5. Process improvement: SPC software can help to identify areas where process improvement is needed and can provide tools for implementing and tracking improvement efforts.

Overall, the use of software can help to streamline the SPC process and improve the accuracy and effectiveness of the analysis. This can help organizations to identify and correct process problems more quickly, leading to improved quality and productivity.

At IntellaQuest, we are beyond excited to introduce our newest application, InspectionQuest, to help streamline the SPC process for manufacturers. We offer a free 60-day proof of concept (POC) for our customers to help them evaluate the effectiveness of our products – If you are interested in learning more about InspectionQuest or taking advantage of our free POC offer, reach out so you can see for yourself the difference our solutions can make in improving the collection and analysis of inspection data and the impact that the insights will have on your production processes.


Key Inspection Challenges Addressed

  • Lack of transparency, consistency and discipline in collecting and analyzing inspection data
  • No organized or scheduled plans of inspection to collect consistent inspection results
  • No defined or standardized inspection techniques or sample sizes
  • No dashboards to see current inspection data, trends, and capability for key processes
  • Creating excess scrap and rework by identifying and addressing out-of-spec conditions too late
  • The same nonconforming situations happening over and over again
  • Inspection data and records scattered on paper checksheets throughout the plant
  • Manual approval or lack of approval with poor documentation for inspection data captured on paper checksheets on the plant floor
  • Cross-functional collaboration to define process improvement items to address out-of-spec or out-of-control problems is difficult
  • Manual tracking of tasks which get “lost” over time with poor accountability
We use cookies to improve your experience on our site, and to keep it reliable and secure.
To find out more, please read our Privacy Policy.