Multi-response optimization (MRO) is a statistical technique that can improve the performance of machines by considering multiple responses simultaneously. By optimizing machine parameters, manufacturers can increase productivity, reduce expenses, and improve product quality. There are many different methods for solving MRO problems, but the Technique for Order Preference by Similarity to the Ideal Solution (TOPSIS) is one of the most used. Hence, this paper presents a novel TOPSIS linear programming model based on the Taguchi method for solving the MRO of a fish scale scraping machine in order to improve fish scaling removal efficiency while reducing fish damage. The experiments were initially designed using the Taguchi experimental design. The proposed TOPSIS model was then used to convert the MRO problem into a single objective optimization problem. Finally, the Taguchi method based on the results of the TOPSIS model was developed, and the optimal combination of parameters (speed = 50 rpm, time = 180 seconds, and capacity = 30 kg) was determined. Compared with the initial parameters, the fish damage decreases by 23.95%, and the fish scaling removal efficiency increases by 29.74%. In addition, it is believed that the TOPSIS linear programming model can be extended to solve other MRO problems.