How do you update and maintain process variation charts over time and respond to changes in process behavior?

1 Choose the right type of chart

The first step is to choose the right type of process variation chart for your data and process. There are different types of charts for different types of data and variables, such as attribute charts for discrete data, variable charts for continuous data, and special charts for non-normal or rare-event data. You should also consider the sampling method, frequency, and size, as well as the process stability and capability, when selecting the appropriate chart. For example, if you are monitoring the number of defects per unit in a stable process with constant sample size, you can use a p-chart; if you are monitoring the diameter of a pipe in an unstable process with variable sample size, you can use an X-bar and R chart.

2 Collect and plot the data

The next step is to collect and plot the data on the process variation chart. You should collect the data from the process at regular intervals, such as daily, weekly, or monthly, depending on the process characteristics and objectives. You should also ensure that the data is accurate, reliable, and representative of the process. To plot the data, you should calculate the average and the standard deviation of each sample, and plot them as points on the chart. You should also calculate and plot the center line, which represents the overall average of the process, and the upper and lower control limits, which represent the expected range of variation within which the process should operate.

3 Analyze the chart

The third step is to analyze the chart and look for any signs of changes in process behavior. Compare the data points to the center line and the control limits, and check for any patterns or anomalies that could indicate the process is out of control or unstable. These may include a point outside the control limits, a run of seven or more points on one side of the center line, a trend of six or more points steadily increasing or decreasing, a cycle of five or more points above and below the center line, or an outlier that is very different from the rest of the data. All these could suggest that the process has experienced a significant variation, shifted or drifted from target value, has a systematic or gradual change over time, has a periodic or seasonal variation, or an unusual event.

4 Identify and eliminate the causes

The fourth step is to identify and eliminate the causes of the changes in process behavior. You should investigate the possible sources of variation, such as changes in materials, methods, machines, measurements, or environment, and determine which ones are responsible for the observed patterns or anomalies. You should also distinguish between common causes and special causes of variation. Common causes are inherent in the process and affect every output; special causes are external or random factors that affect only some outputs. You should aim to reduce or eliminate the special causes of variation, as they are unpredictable and undesirable. You can use tools such as fishbone diagrams, Pareto charts, or 5 whys to help you identify and eliminate the causes.

5 Implement and monitor the improvements

The fifth step is to implement and monitor the improvements. You should apply the appropriate solutions or actions to address the causes of variation and improve the process performance. You should also monitor the effects of the improvements on the process variation chart and check if they have achieved the desired results. You should document the changes and update the center line and the control limits accordingly. You should also continue to collect and plot the data on the chart and analyze it for any new or recurring changes in process behavior.

6 Review and revise the chart

The final step is to review and revise the chart periodically. You should evaluate the effectiveness and relevance of the chart and make any necessary adjustments or modifications. You should also consider if you need to change the type of chart, the sampling method, frequency, or size, or the control limits, depending on the changes in the process or the data. You should also review and revise the chart whenever there is a significant change in the process, such as a new product, a new customer, a new supplier, or a new regulation. By reviewing and revising the chart regularly, you can ensure that it reflects the current state and performance of the process.