What is a Control Chart?
A Control Chart is a powerful statistical tool used to monitor, control, and improve process performance over time. Developed by Walter A. Shewhart in the 1920s, control charts help distinguish between common cause variation (natural fluctuations in a process) and special cause variation (unexpected disturbances). This enables organizations to maintain process stability and achieve consistent quality output.
At its core, a control chart plots process data points over time along with a center line (usually the process mean), an upper control limit (UCL), and a lower control limit (LCL). If the process stays within these limits, it’s considered “in control.” Any data point outside the control limits, or unusual patterns within the limits, signals a potential issue that should be investigated.
Types of Control Charts
Control charts are broadly classified based on the type of data: Variable Data (measured) and Attribute Data (counted). Below are the most commonly used control charts:
1. XÌ„ and R Chart (Mean and Range Chart)
Used to monitor the mean and range of a process when the sample size is small (typically 2-10). It’s ideal for measuring consistency in dimensions, weight, or time.
2. XÌ„ and S Chart (Mean and Standard Deviation Chart)
Used when the sample size is larger (more than 10), providing a better understanding of process variation using standard deviation.
3. I-MR Chart (Individuals and Moving Range Chart)
Suitable for processes with a sample size of 1. Common in service or healthcare industries where frequent sampling is not feasible.
4. P Chart (Proportion Chart)
Used to monitor the proportion of defective items in a process when the sample size varies. It’s common in quality inspections.
5. NP Chart (Number of Defectives Chart)
Used when the sample size is constant, and you are counting the number of defective items.
6. C Chart (Count of Defects per Unit)
Tracks the number of defects in a single unit when opportunities for defects are consistent.
7. U Chart (Defects per Unit Chart)
Used when the number of defects is recorded across varying sample sizes.
Examples of Control Chart Usage
- Manufacturing: Monitoring the diameter of bearings using an XÌ„ and R chart to ensure they stay within tolerance.
- Healthcare: Tracking patient wait times in a hospital using an I-MR chart.
- Call Centers: Using a P chart to observe the proportion of calls dropped each day.
- Food Industry: Using a C chart to count the number of defects in food packets per batch.
- Software Development: Using a U chart to track defects per lines of code in software releases.
Conclusion
Control charts are essential tools in the Lean Six Sigma toolbox, providing real-time insights into process behavior and enabling proactive decision-making. By choosing the right type of control chart based on your data, you can identify trends, maintain process stability, and achieve consistent quality. Whether you’re in manufacturing, services, or software, implementing control charts can significantly improve your operational performance.