Quality Control Chart For Poisson INAR (1) Process

Statistical process control is a theory technique of using statistical methods to monitor the production process. Its core idea is prevention and the most important tool is the control chart. Through anglicizing and tracing to the inspection data in the production process to distinguish abnormal fluctuations, in order to timely warning to reducing defection, and to improve production efficiency. But the traditional statistical process control theory is based on a basic assumption: the data independence of process observation. In actual production, the process data is not always satisfied with the assumption of independence, which makes some control charts are no longer suitable for the process of correlation.In this paper, we study the poisson INAR(1) process. Firstly, combining the AR(1) process with the poisson counting process, due to the integer value of poisson counting pocess, we must correct the original model. In this paper, we use two methods to correct the model. Based on the new model, the control limits of c-chart and residual chart are constructed, in order to adapt the new model.In addition to the modification of the model and the control chart, the paper compares the two control charts under different conditions. Taking the average run length as the criterion, the simulation is carried out under controlled and uncontrolled conditions. Comparing the effect of the control charts, we know the most suitable chart under different conditions. In the end of this paper, the chart of average run length is discussed, and the relationship between the average run length and other variables is observed.