WSEAS Transactions on Systems and Control
Volume 2, Issue 2, February 2007, Pages 193-200

Consistency comparison of normalization procedures in multiattribute decision making

 
Clayton School of Information Technology, Faculty of Information Technology, Monash University, Clayton, Victoria 3800, Australia


Abstract

Multiattribute decision making (MADM) often uses a normalization procedure to transform performance ratings with different data measurement units into a comparable unit. Although there are a few normalization procedures available, MADM methods usually use one of these procedures without justifying it. This study performs a consistency comparison among four commonly used normalization procedures when used to solve MADM problems with varying number of alternatives. The simple additive weighting (SAW) method, the most widely used MADM, is used in this experiment for the aggregation purpose. The validation of the results using different normalization procedures is examined using two performance measures: ranking consistency and overall preference value consistency. A new simulation technique is used to generate MADM problems with various numbers of alternatives. The result of the simulation study shows that vector normalization and linear scale transformation (max method) are most consistent when used with SAW.

Keywords: Multiattribute decision making, Normalization, Simulation, Method comparison, Decision consistency