Sewer networks play a vital role in urban infrastructure, but the formation of hydrogen sulfide poses challenges due to its odor and corrosion effects. Traditional models have limitations in predicting this complex data, leading to the exploration of boosting machine learning. By using over 700 datasets and 11 parameters, including pH, dissolved oxygen, temperature, and sulfate concentration, a new XGBoost-based model was developed to predict hydrogen sulfide formation in sewer systems. The model outperformed other algorithms with high prediction efficiency and accuracy, offering real-time sulfide control strategies. This innovative approach can help minimize sulfide generation and improve sewer system operations. Hydrogen sulfide production is crucial to monitor due to its impact on corrosion, odors, and health risks in sewer networks. By employing advanced machine learning techniques like XGBoost, this study provides a promising solution to tackle sulfide formation in sewer systems and enhance environmental sustainability.