Evaluation of the Impact of Selected Weighting Methods on the Results of the TOPSIS Technique

Abstract

The present research deals with selected multi-criteria methods identified according to available literature sources as a suitable instrument for the comprehensive evaluation of a set of alternatives. Further, it focuses on a group of methods including Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS), which is often associated in practice with the use of several subjective and objective methods for the determination of the weights of selected indicators. Accordingly, five of these objective methods are used in this research. The main contribution consists in emphasizing the differences between the results achieved through various methods for the determination of the weights of input indicators. Moreover, we restate the fact that the choice of an adequate method to weigh indicators significantly affects the overall results of the TOPSIS technique. The Coefficient of Variance method clearly identifies subjects in prominent positions. Subsequently, the Mean Weight method does not consider the structure of data and their variability, while focusing on all the indicators being equal. The results obtained via the Standard Deviation and the Statistical Variance Procedure methods are comparable with the results obtained for the identical weights of individual indicators, i.e., the Mean Weight method. Therefore, based on the overall results of our research for the determination of the weights of input indicators for the purposes of the TOPSIS technique, we recommend the use of the Standard Deviation method.