This article aims to use the improved entropy weight TOPSIS evaluation model to conduct decision analysis on the content of the intelligent audit system for digital economy enterprises driven by artificial intelligence in China, in order to solve the current dilemma of internal audit in China and explore new ideas for the development of internal audit. By using literature analysis and expert survey methods, a performance evaluation system for internal audit of enterprises was constructed, including four primary indicators and 19 secondary indicators. The qualitative indicators were quantified using the semantic difference membership degree assignment method, and the quantitative indicators were normalized using the hierarchical processing method. Using entropy method (EWM) to determine objective weights, combined with Lagrange multiplier method to calculate combined weights, and then using improved TOPSIS evaluation method for decision analysis. Taking the audit evaluation of the quality of scientific and technological development in Shandong Province from 2014 to 2024 as an example, the empirical results show that the improved TOPSIS evaluation method can effectively overcome the influence of subjective factors in the decision-making process, reflect the importance of various factors in the evaluation, and the operation is feasible and effective. Compared with the factor analysis method and the comprehensive scientific and technological progress rate of Shandong Province announced by the Shandong Provincial Department of Science and Technology, the ranking order of the comprehensive scores obtained by the improved TOPSIS evaluation method is consistent, which verifies the scientific and applicability of this method. Improving the TOPSIS evaluation method can achieve a transition from qualitative to quantitative, from local to comprehensive, and enhance the systematic and scientific nature of decision-making, providing assistance for content decision-making in intelligent audit systems for digital economy enterprises driven by artificial intelligence. However, the decision-making process is still inevitably influenced by subjective factors, and attention should be paid to the design of given strategies and plans to avoid limitations.
