Quality Control FAQ


Quality Control & Data Measurement

Creating a quality control plan for your system

For some customers their applications are critical to the point where they need to know the cleanliness at point of use. A larger number of tests and/or longer test times will ensure a more representative result (See “What factors can effect particle concentration and distribution in my system).

It is however important to monitor trend over time, and make a fair appraisal so that the right courses of action can be taken to keep system quality at the right level. If you require point of use results, this trending can be done as part of commissioning of your equipment so that you are informed from day one.

Below is a simple statistical tool which can help you in achieving this…


How to get the most from your data?

Normally pre-control charts are used to monitor information from systems. Acceptable cleanliness, or moisture level should be set at a limit well within the upper control limit (Red alarm) so that the system always performs the way it is intended. It is also recommended to use the detailed counts when analysing data, as this can be used more accurately and with greater flexibility.

When analysing your data we would recommend working to at least 4 standard deviations from the mean (Amber alarm) providing a 99.3% confidence interval when predicting the next result. If you are analysing data over the course of 1 day, try to take data points at set intervals throughout that working day to take into account any change in system distribution.

What is the standard deviation?

In statistics and probability theory, standard deviation shows how much variation or “dispersion” exists from the average. A low standard deviation indicates that the data points tend to be very close to the average, whereas high standard deviation indicates that the data points are spread out over a large range of values.

The standard deviation is the square root of its variance for a data set. A useful property of standard deviation is that, unlike variance, it is expressed in the same units as the data.

Why is it important to know my system capability?

As discussed in previous sections of this knowledge centre, most systems are heterogeneous in nature. The contaminants in your process are not evenly distributed and therefore the data can vary from instrumentation one minute to the next.

Process capability is the ability for a system to maintain a set working level. In most instances, averages along with the standard deviation need to be taken into account to arrive at a repeatable and predictable result. Taking the average result in isolation can be a source of error.

Knowing the capability of your system can inform you about decisions to replace filters or add filters to your system. Depending on how efficient your system is, it may extend or shorten the time between element replacements. In addition to this, you can use this statistically sound data to help you in other areas of continual improvement and proposal justification. It also makes for more suitable warning alarm limit settings.