A fuzzy approach to qualitative cross impact analysis

This paper proposes a fuzzy approach to qualitative cross impact analysis. Cross impact analysis is seeking to find both direct and indirect relationships of variables relying on experts' decisions to structure and formalize judgmental forecasting. The knowledge is often uncertain or fuzzy when dealing with future events. In addition, experts prefer to use linguistic terms or fuzzy values in their predictions. Therefore, a qualitative cross impact analysis is represented in terms of fuzzy relationships. Four different approaches including crisp binary, crisp rated, fuzzy linguistic and fuzzy rated are applied to a specific case-the security appliances sector. The results are then compared depending on the variables' characteristics. The fuzzy approaches reveal different results than the crisp ones. The fuzzy rated approach makes it possible to "infer in a wider perspective" from the results and pick out hidden variables. On the other hand, the results of the fuzzy linguistic approach help in deciding for variables where indecision is high in other approaches. Finally, in contrast to crisp approaches, the fuzzy approaches are more successful in representing uncertainty.