What I mean by a "running number" is that the average is always changing. If I turn in 10 homework assignments and the grades are as follows: 58, 87, 99, 90, 92, 95, 89, 90, 95, 91 then my average is 88.6. If I turn in 10 more assignments my average changes, we don't say that my next 10 are either in control or out of control. Although, personally, I would have said the 58 was out of control from the start. At what point in a production environment does the average change? My company has a lot of returns, so if I were to start taking data I'm sure I would be calculating an average that contains parts that are below quality standards. Hopefully we would improve, but I don't know when I would create a new baseline, surely we wouldn't want the same baseline forever.
If by "out of control" you mean deviant, I would agree.
Baseline is not the same as average. It is the cutting point at which you will reject the product. It would depend on the quality standards you have for your product.
An an analogy, let's use academic grades. "C" would be average, but between "D" and "F" would be my baseline for rejection.
In a production environment, the average can change depending on various factors such as changes in process, equipment, materials, or human factors. To determine when to create a new baseline or update your average, you can use statistical process control (SPC) techniques.
SPC involves monitoring and controlling a process to ensure it operates within acceptable limits. One commonly used tool is a control chart, which visually displays process data over time. Control charts help identify when a process is experiencing common cause variation (expected variation within the process) or special cause variation (unusual variation due to specific, identifiable factors).
To establish a baseline, you would collect data on a regular basis and plot it on a control chart. The control chart would typically have a centerline representing the average and control limits calculated based on historical data. These control limits indicate the range within which the process is expected to operate.
If the data points consistently fall within the control limits over a sufficient period of time, the process is considered stable, and you can use the average as your baseline. However, if any data points fall outside the control limits or show non-random patterns, it suggests special cause variation, indicating a need to investigate and potentially take corrective action.
In your case, if you start monitoring returns, you would collect data regularly and plot it on a control chart. Initially, you may observe the average including returns below quality standards. However, if you implement improvements and monitor the process over time, you would expect to see less variability and movement towards a higher average. Regular analysis of control charts can help you identify when it is appropriate to update the baseline and assess process performance.
Remember, SPC is a continuous improvement approach, and the goal is to reduce variation and improve the average over time.