Cloud Delphi framework under asymmetric uncertainty

The present work formalizes and establishes a framework for processing expert information where expert responses are not necessarily symmetric. A generalization of the Cloud model is presented and applied during the execution of Delphi rounds. In contrast to conventional approaches, this method captures the cognitive anisotropy intrinsic to expert estimation through the use of split-entropy. Furthermore, a hybrid evolutionary operator is proposed to control inter-round hyper-entropy, taking into account not only opinion variability but also the experts' self-assessment regarding their confidence level in the issued judgment. Boundary conditions and system stopping criteria are also formalized using infinitesimal entropic variation metrics, thereby guaranteeing the stability and robustness of the synthesized consensus. The results are implemented in a real-world scenario, and the data are processed and visualized using a script programmed in the R statistical language.