In the context of digital development, aesthetic education in colleges and universities is encountering novel challenges and opportunities, with some institutions grappling with the issue of an uneven distribution of digital resources for aesthetic education. In order to achieve rationalization and maximization of the benefits of digital aesthetic education resource allocation in colleges and universities, a digital aesthetic education resource evaluation index system and a multi-objective function model of the utilization efficiency and allocation efficiency of digital education resources are constructed. The deep Q network algorithm, which is based on a genetic algorithm, is also used to do model solving and simulations to find the best way to set up digital art education resources in the colleges and universities that are being studied. The simulation results demonstrate an improvement in the comprehensive utilization efficiency of digital aesthetic education resources in each university, ranging from 4.28614 to 12.89167 to 9.95566 to 12.92808, and a tendency towards equilibrium. This improvement not only optimizes the allocation of aesthetic education resources in colleges and universities but also validates the accuracy and usability of the model. This study holds significant importance in enhancing the efficiency of educational resource usage, mitigating disparities between colleges and universities, achieving the set targets, minimizing resource waste, and fostering the advancement of education.