A Novel Analytic Hierarchy Process Technique for Large and Fuzzy Criteria Decision Making Problems

Abstract

This dissertation is a study of the Analytic Hierarchy Process (AHP) and is divided into two main parts. In the first part, the researcher requires the development of a new comparison procedure of an analytic hieratical process to make it convenient to use the AHP analysis to apply on cases with large criteria. The proposed AHP and the scoring methods will be improved to make it simple for experts. The method is called “Normalize function-based scaling AHP” The researcher proposed a novel technique by borrowing the idea of the Likert scale but employing a 1 to 9 scale. By comparing the proposed method with the classic AHP with a clustering technique, the proposed method yielded the same conclusion as the classic AHP while requiring significantly less effort.

Furthermore, the threshold of decision changing was not a substantial discrepancy. In the second part, this research wants to increase the performance of FAHP methods. It is to compare 2 decision-making methodologies, classic AHP and FAHP (Triangle, Trapezoidal) in the case of choosing the preferable medical devices using the weighing results and consistency ratio values on the same data in the case of medical device suppliers. The result, in case, one needs the calculation with less bias, a user should consider FAHP (Triangle) method, as FAHP (Triangle) allows the user to detect and analyze consistency ratio more rapidly but one must accept that it involves more complicated calculation which is considerably recommended for the amateur assessor with an authority to approve such vendor, while classic AHP is suitable for assessors with excessive experience.

-