After obtaining MA control limits, the MA error detection is investigated by simulating detection of systematic error by MA. This is performed by introducing various systematic errors in the dataset at various positions. After this, the error detection performance of multiple MA procedures is investigated. The readout of the simulation is the number of generated assay results until MA error detection. The obtained results are presented in bias detection curves. Here the median number of assay results required is presented for multiple MA procedures and multiple simulated errors.
Bias detection curves can be used to compare the error detection performance of various MA procedures in order to obtain the considered most optimal MA.
MA Generator performs the simulation procedure and generates bias detection curves.