Based on the concept of “user-centered”, this paper designs a product form optimization model based on ant colony algorithm. Through mining the online reviews of the products, we determine the perceptual imagery of users, and categorize the perceptual imagery and determine the weights from the perspective of user satisfaction. Combining the factor analysis of perceptual imagery and the contribution value of morphological features on perceptual imagery, the product morphology optimization fitness function is constructed. Solve the model according to the basic principle of ant colony algorithm, and study the decision-making method to assist product optimization. Take a brand A model forum word-of-mouth data as an example to analyze, obtain users’ perceptual imagery through SO-PMI algorithm, and assign values to perceptual intention weights with the help of cluster analysis. Determine the contribution value of morphological features through the SD investigation of product morphological differences. Genetic algorithm is introduced to carry out comparative experiments to verify the superiority of ant colony algorithm in optimizing model solving. Finally, the application effect of the predictive model solving scheme is analyzed through user satisfaction survey. The results show that the output of the product optimization design model based on ACO algorithm Model A is 8. 23.11% of the users are very satisfied with the optimized Model A, 65.55% of the users are satisfied, and 85.72% of the survey respondents are very willing and ready to buy the optimized Model A.