Blog: Applications

Using Standard Deviation to Dynamically Update LIMS Test Limits in QC Testing

28th July 2021

SD Chart 200x120

Standard Deviation - Revision

Scientists are taught early on to think about the variability and precision of results. This is often discussed mathematically in terms of a normal distribution around the average or mean, together with the standard deviation (σ, called sigma) of the values from the mean. In a normal distribution 68% of the values are within 1σ of the mean (μ), 95% within 2σ, and 99.7% within 3σ.

SD Chart
  • SPC Chart1

We can view these published limits, along with historical limits, by clicking the ‘View Published Limits’ button. The newest version (version 4 in the graphic) of the magnesium limits shows the new mean and standard deviation numbers, along with a historical record of previous limits for audit purposes. This screen highlights the change of numbers over time depending on the results obtained.

Looking at the test maintenance screen (below) we can see that the new 2σ and 3σ limits are now set against the magnesium test component. Future tests will use these new warning limits until the above process is repeated.

Using historical results to dynamically update test limits is a powerful feature to provide a warning of possible erroneous results and process drift in a quality control laboratory. This methodology can be used across manufacturing and environmental QC labs and is particularly prevalent in water QC laboratories.

SD Limits2
  • Test Limits Updated
  • Chart Limits1 3