Absolute order-of-magnitude reasoning applied to a social multi-criteria evaluation framework

A. Afsordegan, M. Sánchez, N. Agell, J. C. Aguado, G. Gamboa

Research output: Contribution to journalArticleResearchpeer-review

5 Citations (Scopus)


© 2015 Taylor and Francis. A social multi-criteria evaluation framework for solving a real-case problem of selecting a wind farm location in the regions of Urgell and Conca de Barberá in Catalonia (northeast of Spain) is studied. This paper applies a qualitative multi-criteria decision analysis approach based on linguistic labels assessment able to address uncertainty and deal with different levels of precision. This method is based on qualitative reasoning as an artificial intelligence technique for assessing and ranking multi-attribute alternatives with linguistic labels in order to handle uncertainty. This method is suitable for problems in the social framework such as energy planning which require the construction of a dialogue process among many social actors with high level of complexity and uncertainty. The method is compared with an existing approach, which has been applied previously in the wind farm location problem. This approach, consisting of an outranking method, is based on Condorcet's original method. The results obtained by both approaches are analysed and their performance in the selection of the wind farm location is compared in aggregation procedures. Although results show that both methods conduct to similar alternatives rankings, the study highlights both their advantages and drawbacks.
Original languageEnglish
Pages (from-to)261-274
JournalJournal of Experimental and Theoretical Artificial Intelligence
Issue number1-2
Publication statusPublished - 3 Mar 2016


  • Absolute order-of-magnitude
  • energy planning
  • linguistic labels
  • multi-criteria decision analysis
  • qualitative reasoning


Dive into the research topics of 'Absolute order-of-magnitude reasoning applied to a social multi-criteria evaluation framework'. Together they form a unique fingerprint.

Cite this