Introduction to Statistical Process Control Charts (Lean Six Sigma)

This paragraph introduces the concept of process control charts, which are tools used to determine if a process is stable over time or if a change has been effective. It explains that these charts have a y-axis representing a critical to quality (CTQ) measurement, which is monitored over time on the x-axis to assess stability or improvement. Examples of CTQ measurements are given, such as flight distance for paper airplanes, battery life for phone batteries, page load time for websites, and delivery time for packages. The paragraph also describes the elements of a control chart, including control limits, process mean, and data points, and distinguishes between control limits and specification limits. It further explains the difference between special cause and common cause variations in a process.

The second paragraph delves into the different types of control charts used for continuous data, such as the Individual Moving Range (IMR) chart, which is suitable for low-volume production, and the X bar R and X bar S charts for higher volume production. The R chart represents the range of a subgroup, while the S chart uses standard deviation. For attribute data, the paragraph discusses four common types of charts: C for constant sample size defects per subgroup, U for variable sample size defects per unit, and NP and P charts for tracking defective units with constant or variable sample sizes, respectively. The paragraph also provides mnemonic devices to remember the types of charts and their applications.

This paragraph discusses the use of tests to detect special cause variation in control charts. It mentions Walter Shewhart, known as the father of Statistical Process Control (SPC), who developed these tests. The paragraph outlines eight common tests used for this purpose, each with specific criteria for identifying out-of-control points or trends in the data. These tests include checks for points beyond three standard deviations from the mean, sequences of points on one side of the centerline, trends in increasing or decreasing points, and patterns of points within one sigma of the mean. The paragraph emphasizes that while SPC can indicate when a process is out of control, it does not explain why, necessitating further investigation.

The fourth paragraph applies the concept of control charts to the real-world scenario of an e-commerce website's loading times. It highlights the importance of fast loading times for customer retention and sales, citing statistics on user expectations and the impact of slow load times. The paragraph describes a hypothetical scenario where load times are measured over 15 days, with the data suggesting the use of an X bar R or X bar S chart due to the high volume of measurements. It also discusses the use of Minitab software to generate control charts and the importance of understanding when a process is out of control, as indicated by the tests, to investigate and improve website performance.

The final paragraph explores the use of control charts to track medication errors in a hospital setting, focusing on defects per unit, or errors per patient. It underscores the significance of reducing medication errors for patient safety and cost reduction. The paragraph presents a scenario where errors are monitored over 25 weeks with variable sample sizes, suggesting the use of a U chart for attribute data with variable sample sizes. It also discusses the process of creating a U chart in Minitab, selecting appropriate tests, and interpreting the results to identify and address points above the control limit, aiming to improve the hospital's error rate over time.