Chang e-Point Estimat on of H igh-Yield Proce sses in the Pres ence of Serial Correlation

Change- point estimation is as neffective method for identifying the time of a change in product on and service proces ses. In most of the statistical qua lity control literature,it is usually assumed that the quality characteristic of interest is independently and identicall distribute d over time. It is obvious that this assumption could be easily vio lated in practice. I n this paper, we use maximum likelihood estimation method to estimate when a step change occurs in high-yield process by allowi ng a serialcorrelation between observation s. Monte C arlo simul tion is used as a vehic le to evaluate the performance of the proposed method . Results indicate sa tisfactory performance for the proposed method.

Keywords:

Change point estimation , High yield process , Maximum likeli hood estimation , Correlation , Statistical process control