Implementing of ahs for process monitoring evaluation system

This research is about implementing of Analytical Hierarchy System (AHS) for process monitoring evaluation. Multivariate Statistical Process Monitoring (MSPM) system is an observation system to validate whether the process is happening according to its desired target. It will detect and diagnose the abnormality of the process behaviour and maintain consistent productivity by giving an early warning of possible process malfunctions. A significant development in MSPM has led to the introduction of principal component analysis (PCA) for reduction of dimensionality and compression of the historical operational data prior to the MSPM’s two statistics which are Hotelling’s T2 and SPE models are used. This paper presents about developments of AHS in PCA-based multivariate statistical processes monitoring (MSPM) system. The procedures in MSPM system consists of two main phases basically for model development and fault detection by using Matlab. This research will be focused on implementing of AHS by using Microsoft Excel for AHS part. From the MSPM framework, the fault identification may trigger the results in contribution plot, SPE statistics and T2 statistic models. Fault detection that produced from PCA in terms of contribution plot are then is applied in AHS as a selection tool to rank or make the priorities of the variables involved. The contribution plot produced from fault identification will be implemented in AHS parts. Normally, decision making involves the following elements which are the decision makers, criteria or indicators and decision methodology. Next, AHS will come out with the hierarchy for six types of contribution plots. Comparison is made based on the ranking and types of faults. In the field of decision making, the concept of priority is essential and how priorities are derived influences the choices one makes or decides. So, AHS is used to make the best selection. As a conclusion, it is proven that the proposed system is able to detect the fault as efficient as the MSPM. Thus, it can be the other alternative method in the process monitoring performance. Finally, it is recommended to use data from other chemical processing systems for more concrete justification of the new techniqu