This study aims to develop and evaluate a heuristic-based expert system capable of optimizing training and preventing injuries in wushu athletes through personalized and adaptive recommendations. The research method used an experimental approach with software engineering, integrating Case-Based Reasoning (CBR) and Rule-Based Expert System (RBES) into a single decision support system. Data were obtained from wushu athletes' injury histories, biomechanical data measured by wearable sensors, and interviews and questionnaires with athletes and coaches. The results showed that the system was able to classify athletes' injury risk into low, medium, and high categories consistently with expert assessments, and reduced the rate of recurrent injuries, particularly ankle sprains, by approximately 30% during the implementation period. Furthermore, the system helped improve training efficiency without compromising athlete safety. The conclusion of this study is that the integration of CBR, RBES, and heuristic approaches produces an adaptive, transparent, and effective system as a decision-making tool for training optimization and injury prevention in wushu athletes. Keywords: Expert System, Case-Based Reasoning, Rule-Based Expert System, Injury Prevention