Document Type

Closed Project

Publication Date

Winter 2006

Instructor

Dundar Kocaoglu

Course Title

Decision Making in Engineering and Technology Management

Course Number

EMGT 530/630

Subjects

Multiple criteria decision making, Decision making -- Mathematical models, Estimation theory, Technology -- Management

Abstract

The dissertation presents a new method to aggregate group pairwise comparison judgments under the framework of AHP. The Minimum Distance Method is built upon the Weighted Geometric Mean method with the objective to determine the set of the exponential weights such that the total distance of pairwise comparison matrices and the aggregated matrix is minimized. The method leads itself to using a goal programming package that is commercially available, LINDO, to solve the optimization problem. The second part of the dissertation presents the tests to evaluate the performances of the AHP aggregation methods. A set of data was generated by computer simulation, and another set was collected from a survey. Tests were run to determine the accuracy and group disagreement of the aggregation methods.. Results suggested that the MDM outperforms others in terms of accuracy, but arithmetic mean does for group disagreement.

Description

This project is only available to students, staff, and faculty of Portland State University

Persistent Identifier

http://archives.pdx.edu/ds/psu/23365

Share

COinS