Statistical Analysis Tools in Six Sigma - Control Charts

What are Control Charts?

Imagine you are a Six Sigma practitioner aiming for optimal process performance. You understand the importance of maintaining consistent quality and preventing defects. You also know that ineffective quality control can lead to costly defects, customer dissatisfaction, and wasted resources. But, how can you effectively monitor and control your processes to ensure success? Without a reliable method for monitoring process variations, it becomes challenging to spot issues promptly. Therefore, the problem lies in the lack of visibility and actionable data. Enter control charts, your key to identifying and resolving issues early on.

Control charts are also called Shewhart charts or statistical process control (SPC) charts. Essentially, they are graphical tools useful in quality control and process management to monitor and analyze process variation. In addition, the control charts offer a visual representation of process performance over time. They enable you to assess stability, identify trends, and detect special causes of variation. By using control charts, you gain the power to guide your processes toward optimal performance.

Types of Statistical Process Control Charts

control charts
  1. X-bar and R Charts:

    X-bar (or average) charts track the average value of a process over time.

    R (or range) charts monitor the variability within each subgroup of data.

    These charts are ideal for variable data, such as measurement or continuous data.

  2. Individuals and Moving Range (I-MR) Charts:

    Individual charts plot individual data points, allowing us to observe trends and shifts.

    Moving Range charts show the variation between consecutive data points.

    These charts are useful when you have only one data point per measurement.

  3. p Charts:

    p charts are useful for attribute data, where data is classified into discrete categories.

    They monitor the proportion or percentage of nonconforming items or events.

    These charts are suitable for situations where you have a fixed sample size.

  4. np Charts:

    np charts are similar to p charts but are useful when the sample size varies.

    They monitor the number of nonconforming items or events in a varying sample size.

    These charts are useful when the number of opportunities for nonconformities changes.

  5. c Charts:

    c charts also deal with attribute data but focus on the count of nonconforming items per sample.

    They help monitor the number of defects per unit, where the sample size remains constant.

Fundamentals of Control Charts:

At its core, a control chart is a graphical representation of changes in the process over time. It helps to distinguish between the common factor variations of the process and the main cause variations resulting from specific factors By distinguishing between these two types of variables, control charts help you to understand the stability of the project and identify areas where intervention is needed.

A control chart generally consists of three main components:
  1. Data points: These are individual measurements or observations collected over time from the event under consideration.
  2. Centerline: Also known as process average or target value, it represents the average of data points and serves as a reference for monitoring process performance
  3. Control limits: Upper and lower limits are derived computationally based on design data. These constraints help identify when the process is behaving as expected (within control) or when there are abnormal variations (out of control).

Control Chart Selection and Interpretation:

  1. Define the Objective

    Before selecting a control chart, clearly define the objective of the analysis. Determine whether the aim is to monitor the process mean, variability, or proportions. Understanding the specific quality characteristic to be controlled is crucial for choosing the right type of control chart.

  2. Data Type and Measurement

    Scale Identify the type of data being collected: continuous or discrete. For continuous data, such as measurements or product weights, use X-bar and R-charts for monitoring process mean and variability. For discrete data, like counts or defectives, consider using p-charts or np-charts for monitoring proportions.

  3. Sample Size and Frequency

    Evaluate the sample size and frequency of data collection. If data is collected in subgroups (e.g., samples of size n at regular intervals), consider using X-bar and R-charts. For individual measurements or small sample sizes, consider using an individual/moving range (I-MR) chart.

  4. Data Distribution and Process Stability

    Assess the data distribution and process stability. If the data follows a normal distribution, X-bar and R-charts are appropriate. If the distribution is non-normal or the process exhibits unstable behavior, explore alternative control charts like median charts or exponentially weighted moving averages (EWMA) charts.

  5. Variation Source Identification

    Distinguish between common cause and special cause variation. If the process is stable and variations are solely due to common causes, a control chart will exhibit random fluctuations within control limits. If assignable or special causes of variation are present, the process may be out of control, and specific corrective actions will be required.

  6. Interpretation of Control Chart Patterns

    Learn to interpret the patterns observed on control charts accurately. Look for trends, shifts, cycles, or points outside the control limits. A pattern of points trending upward or downward could indicate a systematic shift in the process, requiring further investigation.

  7. Action and Decision Making

    Based on control chart analysis, determine appropriate actions and decision-making strategies. When special cause variation is identified, investigate the root cause and implement corrective actions to bring the process back into control. If the process is stable and only exhibits common cause variation, focus on continuous improvement efforts to reduce overall variability.

  8. Periodic Review and Continuous Improvement

    Regularly review control charts to monitor process performance and identify opportunities for continuous improvement. Periodic assessment of control chart data helps ensure the sustainability of quality gains and adherence to established process standards.

Understanding Process Variation: A Comprehensive Exploration

  1. Sources of Variation:

    Variation refers to the differences observed in a process or product over time. These differences can arise from various sources. But, broadly, these are categorized into two types: common cause variation and special cause variation.

    a. Common Cause Variation: Common causes are inherent to the process and, therefore, represent the normal, expected variation. This variation occurs even when the process is in control. Further, these factors are often systemic and can be attributed to factors such as equipment variability, operator skills, environmental conditions, or material variations. Common cause variation is random and stable over time.

    b. Special Cause Variation: Special causes are unpredictable, non-random factors that introduce significant deviations in the process. They are typically caused by specific events or factors that are not consistently present. For example Machine breakdowns, operator errors, design changes, or other abnormal occurrences.

  2. Process Stability:

    Process stability is a critical concept in quality management. A stable process is one where the variation is predominantly due to common causes, thus resulting in predictable and consistent performance. Conversely, an unstable process is characterized by the presence of special causes, leading to unpredictable and erratic outcomes.

    a. Identifying Process Stability: Control charts play a central role in assessing process stability. By plotting data points over time on a control chart, we can visually analyze the patterns and trends in the variation. A stable process exhibits data points that fluctuate within control limits, indicating that the process is under control and variations are solely due to common causes.

    b. Reacting to Process Instability: When a control chart indicates the presence of special cause variation, it signals that the process is out of control and requires immediate investigation. Addressing and eliminating special causes is essential to restore process stability and ensure consistent quality.

  3. Role of Control Charts in Distinguishing Variation:

    Control charts are powerful tools that help us monitor processes over time and distinguish between common cause and special cause variations.

    a. Establishing Baseline Performance: Initially, control charts help establish a baseline of the process’s performance by collecting data and determining control limits based on historical variation. This baseline becomes the reference for assessing future process performance.

    b. Detecting Special Causes: Control charts act as early warning systems, detecting any occurrences of special cause variation. When data points fall outside the control limits or display non-random patterns (e.g., trends, cycles, or shifts), it signals the presence of special causes.

    c. Continuous Improvement: By distinguishing between common cause and special cause variation, control charts guide the allocation of improvement efforts. Improvement initiatives target the elimination of special causes to enhance process stability and reduce overall variation.

Control Chart Implementation:

  1. Define the Objective and Scope:

    Before diving into control chart implementation, clearly define the objective of using control charts and the scope of the process under consideration. Also, understand the critical parameters you want to monitor and improve to ensure your efforts are focused and aligned with organizational goals.

  2. Select the Appropriate Control Chart Type:

    Different types of control charts are suitable for different data types and scenarios. Therefore, choose the most relevant chart based on the characteristics of your data. For continuous data, such as process measurements, use X-bar and R-charts or X-bar and S-charts. For attribute data, like defect counts, consider p-charts, np-charts, or c-charts.

  3. Data Collection and Sampling Strategy:

    Collect sufficient data points to establish meaningful control limits and assess process stability accurately. Determine an appropriate sampling strategy based on the frequency and volume of production. More importantly, ensure the data collected is representative of the process variation and captured at regular intervals.

  4. Establish Baseline Data and Determine Control Limits:

    Calculate baseline statistics, such as the mean and standard deviation, for the data collected. Use these values to set control limits on the control chart. Commonly used control limits are ±3 standard deviations from the mean for X-bar and R-charts and ±3 standard deviations for p-charts and c-charts.

  5. Plot Data on the Control Chart:

    Plot the collected data points on the control chart using appropriate software or tools. The X-axis represents time or sample number, while the Y-axis displays the measured values. Connect consecutive data points with lines to visualize trends and patterns.

  6. Analyze the Control Chart Patterns:

    Interpret the control chart patterns to assess process stability and identify sources of variation. Look for specific signals, such as runs, shifts, and trends, which might indicate the presence of special cause variation. Understanding these patterns will help you distinguish between common cause and special cause variation.

  7. Determine Process Capability:

    Evaluate process capability using control chart data to understand if your process meets customer specifications. For this purpose, calculate process capability indices, such as Cp, Cpk, Pp, or Ppk, to quantify the process performance and identify areas for improvement.

  8. Implement Corrective Actions:

    If special cause variation is detected, investigate the root causes and implement appropriate corrective actions to eliminate or mitigate the issues. In addition, engage cross-functional teams and use quality improvement tools like root cause analysis and DMAIC to drive sustainable improvements.

  9. Monitor and Update Control Charts:

    Continuously monitor the process using control charts to ensure ongoing process stability. Periodically update control limits and reevaluate the chart’s performance as the process improves or changes over time.

  10. Training and Communication:

    Ensure that relevant personnel involved in the process are trained in control chart interpretation and the overall improvement process. In addition, foster a culture of data-driven decision-making and communicate the benefits of control chart implementation to gain organizational support.

By following these step-by-step guidelines, you can successfully implement control charts in diverse industries and processes. Control charts empower organizations to make informed decisions, reduce process variation, enhance quality, and achieve sustainable improvements, thereby contributing to overall operational excellence and customer satisfaction.

When to Use Statistical Process Control Charts?

  • Understand process stability and detect any variations.
  • Monitor process performance and quality levels.
  • Identify sources of variation to improve processes.
  • Set appropriate control limits and target values for the process.
  • Establish statistical control to reduce defects and waste.

Benefits of Using Control Charts

  1. Early Detection of Problems:

    Control charts provide a visual representation of process behavior, and thus, enable you to identify potential issues, deviations, or abnormal patterns.

  2. Data-Driven Decision Making:

    By analyzing the SPC charts, you can make informed decisions based on objective data rather than relying on intuition or guesswork.

  3. Process Improvement:

    SPC charts help identify sources of variation. Therefore, this allows you to target areas for process improvement and implement corrective actions effectively.

  4. Reduced Costs and Waste:

    By detecting and addressing process variations promptly, SPC charts contribute to minimizing defects, waste, rework, and associated costs.

  5. Enhanced Customer Satisfaction:

    Consistently monitoring process quality ensures the delivery of reliable and predictable products or services, leading to higher customer satisfaction levels.

Conclusion

Control charts are valuable tools that provide insights into process performance, stability, and variation. By selecting the appropriate control chart for your specific data and situation, you can effectively monitor processes, make data-driven decisions, and continuously improve quality, leading to better outcomes and customer satisfaction.

Scroll to Top