We present a new measure for evaluating the performance of control charts to detect abrupt changes of finite matrix sequences. The objective is to minimize the probability that the control chart fails to raise the alarm at unknown change point time for a given in-control average run length. We construct and prove the optimal control chart with dynamic control limits in different pre- and post-change distributions. We validate the optimality of the proposed chart by conducting exhaustive experiments on both simulation study and real-world data.

中文翻译：

---

未知变化点处有限矩阵序列的最优控制图

我们提出了一种新的方法，用于评估控制图的性能以检测有限矩阵序列的突变。目的是在给定的控制内平均行程长度下，将控制图未能在未知更改点时间发出警报的可能性降至最低。我们构建并证明了在变化前和变化后分布具有动态控制极限的最优控制图。我们通过对模拟研究和实际数据进行详尽的实验来验证所提出图表的最优性。