A new 2-additive Choquet integral based approach to qualitative cross-impact analysis considering interaction effects

Kadaifci Ç., Asan U., Bozdag E.

TECHNOLOGICAL FORECASTING AND SOCIAL CHANGE, vol.158, 2020 (SSCI) identifier identifier

  • Publication Type: Article / Article
  • Volume: 158
  • Publication Date: 2020
  • Doi Number: 10.1016/j.techfore.2020.120131
  • Journal Indexes: Social Sciences Citation Index (SSCI), Scopus, ABI/INFORM, Aquatic Science & Fisheries Abstracts (ASFA), Compendex, Geobase, INSPEC, Political Science Complete, Social services abstracts, Sociological abstracts, Worldwide Political Science Abstracts, DIALNET
  • Keywords: Qualitative cross-impact analysis, Interaction effect, 2-additive Choquet integral, Fuzzy measure, Quadratic programming, DISCRETE FUZZY MEASURES, FUTURES, DELPHI, AGGREGATION
  • Istanbul Technical University Affiliated: Yes


Cross-Impact Analysis, as one of the most applied futures research techniques, arose from the question of whether interrelationships of future events may provide a basis for forecasting. Over the years this technique has evolved to a major tool for determining variables with highest importance in scenario development in a more effective way. Researchers have discussed certain drawbacks of the technique, especially the need for dealing with interactions (i.e. joint effects of variables). However, no satisfactory solution integrating joint effects into the model has yet been suggested Interaction is an important determinant generally for all systems and particularly for futures research since two supposedly unimportant criteria may have a strong effect in the system when they are considered jointly. In this study, to address this issue, a Qualitative Cross-Impact Analysis based on 2-additive Choquet Integral is developed An example is provided to illustrate the applicability and the effectiveness of this approach. In the example, four different settings are presented for validation purposes and the results are compared to the classical approach. The findings indicate that increasing the weight of the interaction effects along the four settings yields increasingly different results than the classical approach. The proposed approach provides a more realistic representation of the system.