6. Final discussion and future work
In this paper we have proposed Identimod, a new methodology and DSS to model branding problems. Identimod is based on soft computing and Vester’s sensitivity model (VSM) which are appropriate techniques to deal with the intangible variables, uncertainty, linguistic terms, and the complexity that can be found in this type of managerial problem. This paper does not just present the DSS but provides the complete modeling process for branding, divided into four different modules. These modules range from the initial identification of the intangible fuzzy variables to the utilization of the final validated model by marketers and decision makers. We highlighted the importance of involving stakeholders in the brand modeling for its acceptance, and Identimod is therefore designed to support the stakeholders’ discussion and the model agreement in a participatory modeling process through the stages of the four modules. These modules follow an iterative and cyclic process needed to create, diagnose, validate, and simulate branding applications. The DSS incorporates the linguistic definition of variables and effects, along with visual graphs and straightforward simulation tools. In addition, Identimod provides the modeler with specific tools for studying the structure of the model and validating it by means of automatic methods. These methods also represent novel applications of advanced computational techniques; for instance, genetic algorithms (GAs) for automatic calibration of VSM, and social network analysis metrics for key variable detection.