Using Statistical Process Control in Six Sigma Projects

Many businesses struggle to implement quality control successfully. Changes based on intuition alone can be costly and time-consuming. Adopting Statistical Process Control (SPC) enables organizations to make smarter decisions by pointing out areas of inefficiency and suggesting improvement solutions. Using these tools and techniques in Six Sigma allows businesses to switch from reactive to proactive by making data-informed decisions.

Let’s explore how to successfully incorporate this efficient quality control improvement process to reduce costs and improve customer satisfaction.

Key Takeaways:

• Incorporating SPC in Six Sigma enables organizations to switch from reactive problem-solving to proactive quality control, reducing waste and increasing customer satisfaction.
• Control charts and tools like histograms and Pareto charts help distinguish between common and special cause variations, enabling quick corrective actions in production processes.
• Case studies from Motorola and General Electric highlight how integrating SPC with Six Sigma can lead to substantial cost savings, revenue increases, and improved product quality.

Understanding Six Sigma

Six Sigma is a concept of methods, tools, and techniques to improve business processes. It was first developed in 1986 by Bill Smith and Mikel Harry, two engineers at Motorola. The method consists of a data-driven quality control methodology that uses structured techniques to improve processes and identify and eliminate causes of variations. Shortly after the development of Six Sigma, General Electric adopted the process and has since reported a cost savings of over $1 billion and a revenue increase of $12 billion.

Since its development, Six Sigma principles have been widely adopted across industry sectors and successfully merged with the Lean manufacturing methodology, resulting in Lean Six Sigma. Organizations such as Ford Motor Company, American Express, and Xerox have implemented Lean Six Sigma successfully. These companies reported improved customer satisfaction and product quality, reduced costs, and increased revenues.

Six Sigma operates under five core objectives:

• Customer Focused: Prioritize customer expectations during quality control improvements.
• Data-Driven Analysis: Collect and analyze data and measuring performance using SPC, a statistical tool,
• Defect Reduction: Interpret the data-driven analysis results to identify and incorporate variations related to quality improvement or eliminate variations related to quality reduction. 
• Sustainability: Incorporate results into decision-making processes, emphasizing quality improvement for future generations.
• Process Control: Monitor production processes and reduce variations. 

How does Six Sigma Work? 

Six Sigma uses the DMAIC (Define, Measure, Analyze, Implement, Control) model for quality improvement.

• Define the desired process, inputs, and outputs by identifying critical quality characteristics (CQCs).
• Measure the output by collecting data on the CQCs, select an appropriate SPS chart (determined by the type of data measured – continuous or discrete – and type of variation – fixed or varied), and calculate control limits.
• Analyze patterns, trends, or signals of special cause variations (variations exceeding common cause variations that can occur at any point during the process).
• Implement actionable changes to incorporate improvements or eliminate defects in the process.
• Control the quality improvement process by periodically repeating the process, expecting to maintain customer satisfaction and profit.

What is Statistical Process Control (SPC)? 

Statistical Process Control (SPC) is a statistical analysis tool that helps improve quality control processes. It was developed in the 1920s by Dr. Walter A. Shewhart of Bell Laboratories. The goal of SPC is to improve industrial manufacturing and was later introduced to organizations and engineers in Japan.

Organizations use Statistical Process Control (SPC) to improve processes, meet customer expectations, and increase profits. SPC uses various statistical tools to analyze data to better understand how processes behave, helping improve quality control and production efficiency.

Over time, all processes will experience some variation. Variations are categorized as either a common cause variation (expected in a process) or a special cause variation (unexpected in a process and should be addressed immediately). SPC monitors these variations in real-time, which helps pinpoint areas for improvement and allows changes to be made without disrupting the workflow.

Key Tools and Techniques in Statistical Process Control

SPC uses a control chart to show how a process changes over time. A control chart records data within set control limits and highlights unusually high or low results, helping distinguish between common and special cause variations. SPC can also use tools like histograms and Pareto charts to analyze data and find variations, scatter plots, and cause-and-effect diagrams to identify relationships and find solutions to problems.

Many factors must be considered when choosing which control chart to use: data type, number of variables, sample size and frequency, data distribution, process characteristics, process stability, and others.

The following table shows some examples of control charts and when to use them:

Chart	Definition	Use Case
I-MR Chart/ I-Chart	A chart that tracks individual measurements and their variability using a moving range.	Suitable for monitoring individual data points where the sample size is one.
X-Bar R	A chart that uses the mean (X-Bar) and range (R) to monitor process variability and stability in small samples.	Used when the sample size is small (typically 2 to 10) to track the average and variability.
X-Bar S Chart	A chart that uses the mean (X-Bar) and standard deviation (S) to monitor process variability and stability in larger samples.	Ideal for larger sample sizes (greater than 10) to track the average and more precise variability.
U Chart	A chart that monitors the average number of defects per unit across varying sample sizes.	Used when the sample size varies and you need to track defects per unit.
C Chart	A chart that monitors the count of defects in a single product unit.	Ideal for tracking the number of defects in an item or area when defects can occur multiple times.
P Chart	A chart that monitors the proportion of defective items in a process.	Used for tracking the fraction of defective products in a sample when the sample size varies.
NP Chart	A chart that monitors the number of defective items in a process.	Suitable for monitoring a fixed number of defective items in a constant sample size.

SPC charts are a valuable addition to a Six Sigma project. Organizations that utilize them allow teams to spend more time on production and error correction and less time on collecting and measuring data.

8 Steps to Implement Statistical Process Control

SPC is a set of tools that use statistical techniques to control a process method. SPC can be implemented by following the steps below:

1. Define the Objective: Clearly define the objective for implementing SPC, such as reducing variability or improving process stability.
2. Identify the Process: Monitor the specific process or subprocess that significantly impacts quality and customer satisfaction.
3. Identify Key Characteristics: Measure the key characteristics that impact output to identify common or special cause variations and determine whether to incorporate or eliminate them.
4. Select Charts: Based on the data type and specific needs, select the appropriate control chart to measure the data.
5. Collect Data: Determine the subgroup size, sampling size, and frequency based on the key characteristics measured.
6. Calculate Control Limits: Calculate the upper and lower control limits, which will allow the SPC method to identify special cause variations that require immediate attention.
7. Plot, Analyze, and Implement: Plot the data on the selected control chart, analyze variations to determine common or special cause variations, and decide on actions to either implement changes or eliminate variations.
8. Monitor and Maintain: Continuously maintain quality control by repeating the process and making necessary adjustments.

4 Primary Benefits of Statistical Process Control

Statistical Process Control (SPC) benefits organizations across industries. Some of the key benefits that help organizations reach their goals include:

1. Reducing Waste

Motorola incorporated SPC in the mid-1980s in response to rising competition, saving the company $17 billion in waste reduction over ten years.

2. Improving Quality

General Electric achieved a $12 billion increase in revenue after implementing SPC to improve quality.

3. Enhancing Efficiency

Ford Motor Company reduced its defect rate by 90% after implementing SPC in the early 2000s, saving the company $300 million in two years.

4. Increase Customer Satisfaction

Using SPC to identify critical processes impacting customer satisfaction, American Express improved customer satisfaction by 20%.

Incorporating SPC into Lean Six Sigma reduces waste and increases efficiency. When teams understand how to implement SPC, they can expect improved quality production and increased customer satisfaction.

Examples of Challenges and Solutions

Implementing SPC can often be accompanied by challenges, whether introducing it into a new or existing Six Sigma project. Here are some of the common challenges:

Example 1: Lack of Understanding

SPC can be a difficult concept to grasp, and when team members do not fully understand the process, it can lead to misinterpretation and incorrect application.

Solution: Promote a culture of continuous learning by providing training options and mentorship opportunities that pair experienced employees with those with less experience.

Example 2: Resistance to Change

Team members may resist incorporating SPC due to a lack of confidence in the technique or a fear of change in general.

Solution: Involve team members in implementing a change management plan that highlights the benefits of SPC and maintain frequent communication throughout the process

Example 3: Quality and Integrity Issues

Poor quality or inconsistent data can undermine the effectiveness of SPC.

Solution: Develop a specific data collection plan with consistent data auditing procedures.

Example 4: Complex Processes

SPC can be difficult to implement successfully with highly complex processes.

Solution: Simplify and break down complex processes into smaller sub-processes and implement them in step-by-step phases.

Case Studies and Real-world Applications

Many organizations have implemented SPC in Six Sigma to increase profitability and productivity. 

Case Study #1: General Electric (GE) used SPC in the late 1990s to improve the quality of its products and services. Since implementation, GE has increased revenue by $12 billion and saved on costs by $1.5 billion.

Case Study #2: In the 1980s, Motorola faced increasing competition from Japanese manufacturers. After developing and successfully implementing Six Sigma, Motorola improved product quality, reduced waste, and saved $17 billion over ten years.

Both examples of successful implementation of SPC in Six Sigma encourage results from long-term use.

Conclusion

For decades, Six Sigma has been utilized in organizations such as General Electric and Ford Motor Company to boost revenue and reduce costs. The statistical Process Control tool allows organizations to monitor variations in production processes that could either help improve processes or severely harm processes in real-time, allowing for quick analysis and resolution to correct errors.

This quick, real-time process monitoring allows organizations to improve quality control, meet and exceed customer expectations and satisfaction, and reduce waste and costs by identifying potential problems and implementing solutions before they become problematic.

Are you interested in learning more about SPC and other forms of process improvement? Check out our Six Sigma Green Belt or Black Belt course.