A soft computing-based decision support system for multidimensional assessment of non-linear coordinative performance in combat sport athletes using a fuzzy inference system

This research establishes a Fuzzy Inference System (FIS) aimed at modeling, analyzing, and forecasting the coordinative capabilities of male athletes in judo and wrestling, with an emphasis on non-linear variables such as balance, reaction time, and agility. These coordinative elements are intricate and qualitative, necessitating a fuzzy logic methodology for enhanced accuracy in evaluation, as conventional binary assessments (e.g., good/bad) frequently overlook the subtleties of performance. The FIS classifies athletes' capabilities into fuzzy categories (Low, Medium, High), facilitating a more detailed examination of their strengths and weaknesses. By incorporating all physical parameters, it is determined that factors such as LBM (Lean Body Mass), PBF% (Percent Body Fat), BMI (Body Mass Index), and BMR (Basal Metabolic Rate) indicate that the athletes are advancing in both physical conditioning and sport-specific skills. The results imply that the implemented training regimen has a beneficial effect on athletic performance, as shown by positive trends in these parameters. Additionally, the research investigates the incorporation of wearable sensors and biomarkers to improve the accuracy of performance assessments and offer more tailored training protocols. This fuzzy logic framework serves as a robust instrument for optimizing player profiling, talent identification, and training strategies in combat sports, aiding coaches in making better-informed decisions and enhancing overall athlete development.