Quality control is the backbone of any successful manufacturing or service operation. Among the many statistical tools available for monitoring process variation, Shewhart control charts stand out as one of the most effective and widely used methods. Developed by Walter A. Shewhart in the 1920s, these charts help organizations distinguish between normal process variation and signals that indicate something has gone wrong. This comprehensive guide will walk you through everything you need to know about implementing Shewhart control charts in your organization.
Understanding Shewhart Control Charts
A Shewhart control chart is a graphical tool that displays data over time and helps identify whether a process is statistically stable or experiencing unusual variation. The chart consists of three fundamental components: a center line representing the process average, an upper control limit (UCL), and a lower control limit (LCL). These boundaries, typically set at three standard deviations from the mean, define the expected range of normal process variation. You might also enjoy reading about What is Problem Solving with Lean Six Sigma?.
The beauty of Shewhart control charts lies in their simplicity and visual clarity. By plotting data points chronologically and comparing them against established control limits, operators and managers can quickly identify when a process requires intervention. This approach prevents unnecessary adjustments to stable processes while ensuring prompt action when genuine problems arise. You might also enjoy reading about How to Perform the Kolmogorov-Smirnov Test: A Complete Guide for Beginners.
Types of Shewhart Control Charts
Shewhart control charts are categorized into two main families based on the type of data you are measuring.
Variable Control Charts
Variable charts monitor continuous data that can be measured on a scale. The most common types include:
- X-bar and R Charts: Used together to monitor the process mean and range
- X-bar and S Charts: Similar to X-bar and R charts but use standard deviation instead of range
- Individual and Moving Range (I-MR) Charts: Applied when you have individual measurements rather than subgroups
Attribute Control Charts
Attribute charts track discrete data such as counts or proportions. Common types include:
- P Charts: Monitor the proportion of defective items in a sample
- NP Charts: Track the number of defective items when sample size is constant
- C Charts: Count defects per unit when the sample size is constant
- U Charts: Monitor defects per unit when sample size varies
Step-by-Step Guide to Creating a Shewhart Control Chart
Step 1: Define Your Process and Quality Characteristic
Begin by clearly identifying what you want to monitor. For manufacturing, this might be the diameter of machined parts, the weight of packaged products, or the number of surface defects. For service industries, you might track response times, error rates, or customer satisfaction scores. The key is selecting a measurable characteristic that directly impacts quality.
Step 2: Collect Baseline Data
Gather at least 20 to 25 subgroups of data when the process is operating under normal conditions. Each subgroup should contain multiple measurements taken at specific time intervals. Consistency in data collection is crucial for accurate results.
Step 3: Calculate Control Limits
Let us work through a practical example using sample data from a manufacturing process. Imagine a company producing metal rods with a target diameter of 10.00 millimeters. Quality inspectors measure five rods every hour for 20 hours.
Here is a subset of the collected data:
Hour 1: 10.02, 9.98, 10.01, 9.99, 10.00 (Mean = 10.00, Range = 0.04)
Hour 2: 10.01, 10.03, 9.97, 10.00, 9.99 (Mean = 10.00, Range = 0.06)
Hour 3: 9.96, 10.02, 10.01, 9.98, 10.03 (Mean = 10.00, Range = 0.07)
Hour 4: 10.04, 9.99, 10.01, 10.02, 9.94 (Mean = 10.00, Range = 0.10)
Hour 5: 9.98, 10.00, 9.97, 10.01, 10.04 (Mean = 10.00, Range = 0.07)
After collecting all 20 subgroups, calculate the grand average (X-double bar) and average range (R-bar). For our example:
- X-double bar = 10.00 mm
- R-bar = 0.065 mm
Using standard control chart constants (A2, D3, D4) based on subgroup size, calculate the control limits. For a subgroup size of 5:
- UCL for X-bar chart = X-double bar + (A2 × R-bar) = 10.00 + (0.577 × 0.065) = 10.038 mm
- LCL for X-bar chart = X-double bar – (A2 × R-bar) = 10.00 – (0.577 × 0.065) = 9.962 mm
- UCL for R chart = D4 × R-bar = 2.114 × 0.065 = 0.137 mm
- LCL for R chart = D3 × R-bar = 0 × 0.065 = 0 mm
Step 4: Plot the Control Chart
Create a graph with time on the horizontal axis and your measured values on the vertical axis. Draw the center line, upper control limit, and lower control limit. Plot each subgroup mean on the X-bar chart and each subgroup range on the R chart.
Step 5: Interpret the Control Chart
A process is considered out of control when you observe any of these patterns:
- One or more points fall outside the control limits
- Seven or more consecutive points fall on one side of the center line
- Seven or more consecutive points show an increasing or decreasing trend
- Fourteen or more points alternate up and down
- Two out of three consecutive points fall in the outer third region between the center line and control limits
Real-World Application Example
Consider a customer service center monitoring call handling times. The team collects data on five calls every day for 25 days. They calculate that the average handling time is 8.5 minutes with control limits of 6.2 minutes (LCL) and 10.8 minutes (UCL).
On day 23, the average handling time jumps to 11.2 minutes, exceeding the upper control limit. This triggers an investigation revealing that a new software update caused system slowdowns. The team addresses the technical issue, and handling times return to normal. Without the control chart, this problem might have gone unnoticed for weeks, resulting in customer dissatisfaction and lost business.
Common Mistakes to Avoid
While Shewhart control charts are powerful tools, several pitfalls can undermine their effectiveness:
- Using Specification Limits Instead of Control Limits: Control limits are based on actual process performance, not engineering specifications. Confusing these leads to incorrect interpretations.
- Reacting to Every Variation: Points within control limits represent normal variation. Adjusting the process for these points often increases variation rather than reducing it.
- Insufficient Data: Creating control charts with too few data points results in unreliable control limits.
- Ignoring Out-of-Control Signals: The purpose of control charts is to prompt investigation and action. Ignoring signals defeats the entire purpose.
- Failing to Update Control Limits: As processes improve or change, control limits should be recalculated to reflect the new normal.
Benefits of Implementing Shewhart Control Charts
Organizations that properly implement control charts experience numerous advantages:
- Early detection of process problems before they result in defective products
- Reduced waste and rework costs
- Improved process understanding among operators and managers
- Data-driven decision making rather than relying on intuition
- Enhanced customer satisfaction through consistent quality
- Documentation of process capability for regulatory compliance
Taking Your Quality Management Skills Further
Shewhart control charts represent just one component of a comprehensive quality management system. To fully leverage these tools and develop expertise in process improvement, structured training is essential. Understanding the theoretical foundations, mastering the calculations, and learning to interpret patterns requires guidance from experienced practitioners.
Lean Six Sigma methodology incorporates Shewhart control charts as fundamental tools within its Define, Measure, Analyze, Improve, and Control (DMAIC) framework. Through formal training, you will gain hands-on experience with statistical software, learn advanced interpretation techniques, and develop the ability to lead quality improvement projects that deliver measurable results.
Whether you are a quality professional looking to enhance your credentials, a manager seeking to improve departmental performance, or an aspiring process improvement specialist, proper training provides the foundation for success. Certified professionals command higher salaries, enjoy greater career opportunities, and make significant contributions to organizational excellence.
Conclusion
Shewhart control charts have stood the test of time as reliable tools for monitoring and controlling process variation. Their visual nature makes complex statistical concepts accessible to frontline workers, while their mathematical rigor satisfies the most demanding quality professionals. By following the steps outlined in this guide and avoiding common pitfalls, you can implement control charts that drive meaningful improvements in your organization.
The journey to quality excellence begins with education and continues with practical application. As you develop proficiency with control charts and other statistical tools, you will discover opportunities for improvement that were previously invisible. The investment in learning these methods pays dividends throughout your career and transforms your ability to solve complex problems.
Enrol in Lean Six Sigma Training Today and master the statistical tools that industry leaders rely on to maintain competitive advantage. Our comprehensive certification programs provide the knowledge, skills, and credentials you need to advance your career and drive organizational success. Do not wait to start your quality improvement journey. Transform your approach to problem-solving and join the community of professionals who are shaping the future of quality management. Contact us today to learn more about our training options and take the first step toward certification.
