A novel computational intelligence approach to making efficient decisions under parametric uncertainty of practical models and its applications to Industry 4.0

A fundamental issue in statistics is the decision making of expected outcomes under parametric uncertainty based on past and present knowledge. This issue occurs in various situations and has a variety of solutions. In this research, a novel computational intelligence approach with applicability to Industry 4.0 is proposed to make effective judgments under parametric stochastic model uncertainty. Ancillary statistics and crucial variables, whose distributions do not depend on the unknown parameters, are utilized since it is expected that only the functional form of the underlying distributions is defined but that some or all of its parameters are unknown. The unique computational intelligence approach efficiently isolates and removes unidentified factors from the underlying models. The proposed approach is innovative in the theory of statistical decisions since it is independent of prior choice, in contrast to the Bayesian approach, which depends on prior choice. It enables the removal of unknown parameters from the issue and the discovery of effective statistical decision rules, which frequently carry lower risk than any other known decision rules. Examples from the real world are provided to demonstrate the suggested strategy.