Statistical process monitoring (SPM) is a data-driven statistical method for monitoring and controlling an on-going process. Over the years, it has gained significant improvements regarding process monitoring not only in the industrial or manufacturing sector, but also in the business sector, aviation sector, health sector, among others. In this research, some advanced and more relevant SPM schemes are proposed to efficiently monitor different processes. The proposed schemes are grouped into three categories:

The first category discusses different SPM schemes for simultaneous monitoring of both magnitude and time of an event. First, we use a max-function to develop an EWMA-type scheme to simultaneously monitor the magnitude and the time of an event. Secondly, we combine a max-function and a distance-function together to simultaneously monitor the magnitude and time of an event. Both schemes assume that the magnitude and the time of an event are independent. However, there are some cases where the magnitude and the time of an event are not independent. Consequently, we suggest a rate charting scheme that simultaneously monitors an event’s magnitude and time, and assume that they are not independent.

The second category discusses different SPM schemes for monitoring various industrial processes. First, we develop some control chartting schemes to monitor the defective chips on a silicon wafer. The number of defect is assumed to follow a zero-inflated mixture Poisson distribution. The schemes have good performance in detecting an upward shift in the number of defective chips. Secondly, we suggest a CUSUM-type schemes to monitor the failure rate of a vertical boring machine. The failure rate is assumed to follow a Maxwell distribution.

The third category presents different EWMA-type schemes for monitoring the parameters of a Gaussian distribution. The first scheme monitors the location parameter efficiently by incorporating extra information from an auxiliary variable that is positively related to the study variable. We also discuss the case where the auxiliary variable is negatively related to the study variable. Then, we suggest some auxiliary based schemes to monitor dispersion parameter. Finally, the last scheme simulataneously monitors the location and the dispersion parameters of a Gaussian distribution.

The performances of the proposed schemes are then examined. They show a better performance than their existing counterparts. Illustrations of the schemes are demonstrated with both real and simulated datasets.