The study optimized the ventilation strategy in high-altitude tunnel construction through big data analysis and machine learning techniques, and developed a high-precision prediction model. Using a combination of random forest and long short-term memory network (LSTM), the environmental data collected during the initial excavation stage is used to predict the ventilation demand and environmental changes during long-distance excavation. When the tunnel is excavated to 100 m, high-precision sensors are installed to collect real-time data such as temperature, humidity, wind speed, dust, and CO concentration, which are transmitted to the central database for preprocessing through wireless networks. The model optimizes hyperparameters through grid search and random search, and evaluates performance using 10-fold cross-validation. The results show that: The RF and LSTM models achieved prediction accuracies of 89% and 93%, respectively. It has demonstrated outstanding predictive capabilities and practicality in actual construction. When the tunnel was excavated to a depth of 1 km to 2 km, the predicted values of the model were highly consistent with the measured data, with an error rate of less than 7%. It effectively responded to large-scale blasting operations, ensuring air quality and worker safety. Compared to traditional methods, the RF and LSTM fusion model better handles complex nonlinear relationships and time-series data, improving ventilation system design and operational efficiency. This integrated approach offers a significant advancement in managing tunnel construction in challenging high-altitude environments, providing reliable predictions to maintain safe and efficient working conditions.