Search

Considerable effort has been spent on the development of normalization models in multi-attribute decision-making (MADM) but despite all of these there is no definitive answer to question: which technique is the most appropriate? Therefore, after a thorough review of the literature, thirty-one methods were identified, classified and evaluated for use in materials selection problems. The objective of this paper is to examine the shortcomings of normalization methods and suggest ways of improving their use in the engineering design decision-making process. The emphasis is placed on materials selection, for problems that include target criteria, as well as cost and benefit considerations, typically seen in more challenging applications such as aerospace and biomedical engineering. It is shown that although many normalization methods may appear to be minor variants of each other, these nuances can have important consequences in engineering design decision-making. To conclude, some dimensionless methods are proposed. The result of this research investigation will help ensure engineering decision makers in general improve their current use of MADM methods but in particular aid designers in developing suitable design performance indices for materials selection. [ABSTRACT FROM AUTHOR]

Considerable effort has been spent on the development of normalization models in multi-attribute decision-making (MADM) but despite all of these there is no definitive answer to question: which technique is the most appropriate? Therefore, after a thorough review of the literature, thirty-one methods were identified, classified and evaluated for use in materials selection problems. The objective of this paper is to examine the shortcomings of normalization methods and suggest ways of improving their use in the engineering design decision-making process. The emphasis is placed on materials selection, for problems that include target criteria, as well as cost and benefit considerations, typically seen in more challenging applications such as aerospace and biomedical engineering. It is shown that although many normalization methods may appear to be minor variants of each other, these nuances can have important consequences in engineering design decision-making. To conclude, some dimensionless methods are proposed. The result of this research investigation will help ensure engineering decision makers in general improve their current use of MADM methods but in particular aid designers in developing suitable design performance indices for materials selection. [ABSTRACT FROM AUTHOR]