Psychology meets artificial intelligence: A human-centered fuzzy optimization model for complex environmental decision-making

Decision-making in highly complex environments, such as environmental investment planning, often involves uncertainty, competing priorities, and the integration of both technical and human factors. From a psychological perspective, this raises challenges related to cognition, judgment, and the interaction between human experts and computational systems. Addressing these issues, this study introduces a novel human-centered decision-making framework that combines artificial intelligence (AI) with fuzzy optimization techniques to enhance transparency, interpretability, and expert engagement. To ensure methodological coherence, the framework links dynamic multi-facet fuzzy Bayesian networks with dynamic multi-facet fuzzy principal component ranking optimization in a complementary manner: the Bayesian network component models interdependencies and uncertainty patterns in expert cognition to generate psychologically informed criteria weights, which are subsequently incorporated into the principal component optimization stage to maximize discrimination among alternatives and produce a consistent ranking structure. Specifically, the model integrates dynamic multi-facet fuzzy Bayesian networks for determining criteria weights and applies a dynamic fuzzy ranking optimization method based on principal components to evaluate market alternatives. Unlike conventional approaches, the framework emphasizes iterative expert participation, aligning computational intelligence with cognitive processes in decision-making. The findings reveal that performance monitoring emerges as the most critical factor in technical feasibility, while agriculture and mining represent the most influential sectors. Beyond methodological innovation, the study demonstrates how interactive AI-supported systems can strengthen decision reliability, support sustainable choices, and bridge organizational psychology with technology-enhanced cognition. These insights contribute to advancing psychological research on human-computer interaction, judgment under uncertainty, and organizational decision-making.