During the analysis of economic resilience of enterprises and industries within existing scientific approaches based on fuzzy sets and soft computing methodology, a number of problems have arisen that require solutions, namely: a) making adjustments to existing linguistic norms of resilience factors; b) incorporating mechanisms of intelligent filtering when forming industry indices; c) improving the accuracy of decision models embedded in the corresponding strategic map. This article describes a series of problematic cases that can be improved within the proposed approaches, including by: a) transitioning from interval membership functions in the classifier to trapezoidal ones; b) excluding noise-contaminated context from the sample that has been rejected based on formal criteria; c) phasifying decision parameters on the strategic map as part of a 4х6 matrix. In the latter case, the technology of pairwise comparison of fuzzy numbers is applied to assess the risk of inefficiency in the decision-making process. It can be formulated that in solving problems related to fuzzy descriptions, we have proposed a series of solutions using the same fuzzy descriptions, but in a more advanced version. In fact, we suggest upgrading a number of previously proven approaches for analyzing economic resilience within the fuzzy scientific paradigm.