Control Charts
Control charts are used for monitoring the variability and to provide agr aph ical display of statistical control.  Astandard, a reference material ofknown c oncentration, is analyzed at specified intervals (e.g., every 50 samples). The result should fall within a specified limit, as these are replicates.The only variation should be from random error. These results are plottedon a control chart to ensure that the random error is not increasing or that a ystematic bias is not taking place. 

Figure 1.1.Control chart

In the control chart shown in Figure1.1, replicate measurements are plo tted as a function of time. The centerline is the average, or expected v alue . The upper (UCL) and lower (LCL)control limits are the values within which  the measurements must fall. Normally, the control limits are ±3σ, within which 99.7% of the data should lie.For example, in a laboratory carrying out microwave extraction on a dailybasis, a standard reference material is extracted after a fixed number ofsamples. The measured value is plotted on the control chart. If it falls outside  the control limit, readjustments are necessary to ensure that the processstays under control.
Control charts are used in many di¤erent applications besides analyticalm e asure ments. For example, in a manufacturing process, the control limitsar e often based on product quality. In analytical measurements, the contr ollimits can be established based on the analyst’s judgment and the experimental results. A common approach is to use the mean of select measurements as the centerline, and then a multiple of the standard deviation is usedto set the control limits. Control charts often plot regularly scheduled analysis of a standard reference material or an audit sample. These are thentracked to see if there is a trend or a systematic deviation from the centerline.