In modern society, the demand for environmental facilities is increasing, and how to effectively design and plan environmental facilities has become an urgent issue. However, traditional design methods often consider only certain requirements and perspectives, resulting in design results deviating from the expectations of actual users. In this study, first, perceptual fuzzy decision-making and design transfer learning were selected as methods. Second, by applying multiple perspectives to environmental facility design methods, these two methods were combined, and a new joint algorithm was proposed. Third, when designing environmental facilities, a joint processing framework was constructed considering the impact of human factors, environmental parameters, and cultural value parameters on the design results. Last, the proposed joint algorithm was validated for functionality and satisfaction. The experimental results of this article indicate that in temperature control design, the accuracy of this research model is 17.7–19.6% greater than that of traditional centralized algorithms. In terms of lighting design, the model results of this study are good, with an increase of 16.7–20.2%. This method comprehensively considers the various dimensional requirements of environmental facilities and has good migration performance. In future studies, we will further investigate the applicability of this method in different scenarios and applications to promote the further development of environmental design.