ابوذر بازیاری

Today, advances in technology allow engineers to collect large samples of a process or product, thus reconstructing the functional relationship that describes the performance of the process or product. This functional relationship is commonly called a profile. This is the relationship between the quality characteristic of the product or process and the developer variables over time. At each stage of the sampling, the observations include a set of data that can be represented by a curve. Monitoring of profiles can be calculated linearly and nonlinearly based on the type of application. Most control charts for monitoring profiles are based on the fact that the observations in each profile are independent of each other. If in practice this is not the case. Sequential measurements in profiles often show sequential correlations. This dissertation focuses on monitoring phase 2 linear profiles when data are interdependent. The Gaussian process model is used to show the correlation in this project. Two examples of Schohart multivariate control diagrams for linear trend monitoring and intra-profile correlation are presented separately in phase 2, and the proposed approach in this dissertation is compared with alternative methods through numerical simulation in which different profile correlations Intended in control. The simulation study shows that when there is a strong and effective correlation in detecting large changes in the correlation of the profile period, the control diagrams proposed in this dissertation are sensitive to changes in the linear allocation process. Finally, an example is provided to illustrate the proposed control chart in this dissertation.