Improving the entropy weighting method performance by using alternative normalisation procedures

There is no doubt that the Shannon entropy's is the most popular method to obtain attributes weights in multi-criteria decision making (MCDM) area. Despite its usefulness, entropy has some notable limitations. Historically, since Shannon and Weaver introduced entropy concept in 1947, and Zeleny in which applied it for attribute weights in 1976, up to present times, a significant drawback with this approach is that a very important attribute may have little weight in analysis. Because all the alternatives are close in terms of that, attribute. So, in this paper to resolve this limit a new normalized method is considered. The finding of this study reveals that some methods (i.e., Peldschus, Juttler-Korth, Max and Stopp) are better than others (total ratios, vector, Weitendorf's, Voogd ratio, logarithmic and product method).