Journal of Information & Systems Management Lian Qian 16 1 2026 https://doi.org/10.6025/jism/2026/16/1/21-30 https://www.dline.info/jism/fulltext/v16n1/jismv16n1_3.pdf The paper explores the application of the Group Method of Data Handling (GMDH) algorithmenhanced with fuzzy inference to analyze and optimize the value chain in the sports industry. The value chain framework encompasses enterprise relationships, value creation processes, and structural dynamics, all of which are critical to sustainable industrial growth. Given the sports industry's limited historical data and inherent uncertainty, traditional modeling approaches face challenges. The GMDH algorithm offers advantages such as self organization, global optimization, and effectiveness with small datasets. However, its susceptibility to overfitting and structural instability prompted the integration of fuzzy logic to handle imprecise and noisy data better. The fuzzy GMDH model improves prediction accuracy by incorporating fuzzy reasoning into the basic processing units, enabling more robust modeling of nonlinear and uncertain systems. The study applies this optimized model to the sports industry in "M Province," comparing performance between standard GMDH and fuzzy GMDH. Results indicate that the fuzzy variant provides superior fit and predictive capability, supporting more accurate evaluation of the industry's value and development potential. Despite these improvements, the authors acknowledge room for further refinement of the algorithm's architecture. The research underscores the importance of advanced data mining techniques, such as fuzzy GMDH, in enabling evidence based decision making in emerging sectors, including sports, where data scarcity and volatility are common challenges.