Enhancing the productivity and quality of Balinese cattle is a crucial goal for improving livestock management practices in Indonesia. Traditional evaluation methods used by farmers are often subjective and inconsistent, leading to inaccuracies in cattle classification and limiting the effectiveness of breeding and selection processes. To address these challenges, this study proposes a Fuzzy Inference System with Certainty Factor (FIS-CF) to improve cattle classification by providing more objective and reliable grading criteria. The model utilizes key physical parameters, including shoulder height, body length, and chest circumference, as input features to categorize cattle into three quality classes. A diverse dataset was collected from the People's Animal Husbandry School (SPR) and various farms across Indonesia to evaluate the model's performance. The FIS-CF model achieved a classification accuracy of 95.93% and a balanced accuracy of 96.20%, outperforming traditional methods that rely on subjective assessment. These results demonstrate that the proposed model provides a consistent, scalable, and data-driven solution for livestock classification, helping farmers make more informed decisions in cattle selection and breeding. Additionally, the model addresses key limitations of current practices by reducing reliance on manual evaluations, which often vary between assessors. The findings highlight the potential for wider adoption of the FIS-CF model across the livestock sector to improve productivity and streamline herd management processes. Future research will aim to refine the model further by incorporating additional parameters, such as age and weight, and expanding its validation to larger datasets covering different cattle breeds and farming environments to ensure broader applicability in sustainable livestock management.