Increasing discrimination in multi-criteria analysis
Item TypeConference Paper
MetadataShow full item record
Multi criteria methods for supporting decision-making procedures are widely used in sustainability assessment. One of the most important steps in decision-making procedures is the evaluation of policy options or alternatives in order to find a hierarchy of option choices. Utility function value distributions are often constructed for the range of indicators for the options to be assessed. This distribution can be presented as an impact matrix stacked with indicator weights to reflect the relative importance of different indicators for the decision-maker. Solving rules are then introduced to integrate all individual indicator evaluations into a single integral utility estimation. These are often based on the averaging procedures, one of the simplest being arithmetic averaging. Whilst averaging rules are very attractive to decision makers due to their simplicity and logical transparency, using averaging as the first step of the decision-making procedures can significantly reduce the discrimination of the options, especially if there are counteractive individual indicator estimations. This paper proposes a method to evaluate and overcome this loss of discrimination. The paper explains the basis of a discriminatory analysis approach to sustainability assessment demonstrates its application through the use of illustrative data and describes its application to an existing case study where researchers had applied a number of multi criteria analysis tools. It was concluded that the discriminatory analysis provided a useful addition to the decision-makers toolbox as it provided a means of assessment of the validity of the application of the simple arithmetic averaging technique.
Barannik, V., Blackwood, D. and Falconer, R. 2007. Increasing discrimination in multi-criteria analysis. In: M. Horner, et al. eds. 2007. International Conference on Whole Life Urban Sustainability and its Assessment, 27-29 June 2007, Glasgow, UK : Glasgow Caledonian University. [online]. Available from: http://www.sue-mot.org/conference-2007/papers/