In this paper, we used a shewhart T 2 control chart to monitor
a Poisson regression profile in phase II. Note that Yeh et al. [13]
introduced five Hotelling T 2 control charts to monitor binary
profiles in phase I. Any of these T 2 charts uses a different
method to estimate the mean vector and covariance matrix.
They showed that the T 2 control chart, which estimates the covariance
matrix by averaging the covariance estimates of each
given sample, is more effective in detecting both step and drift
shifts. This control chart is applied in phase II with the assumption
that the mean vector and covariance matrix are known. The
T 2 statistic for sample j, j = 1, 2, . . . , T in phase II is defined as: