Air cargo is regarded as a crucial component of modern aviation logistics, ensuring the smooth operation of global supply chains through its safe, efficient, convenient, and high-quality services. Accurate forecasting of air cargo volume is considered to be of significant importance for industry development and supply chain management. However, due to the pronounced nonlinear, non-stationary, and seasonal fluctuation characteristics inherent in air cargo volume data, satisfactory predictive performance is often difficult to achieve using traditional forecasting methods. In this study, monthly air cargo volume data in China from 2009 to 2025 were utilized as the research sample. Based on the construction of three single forecasting models, namely extreme gradient boosting (XGBoost), long short-term memory (LSTM), and Prophet, an extended generalized induced ordered weighted averaging (GIOWA) operator was innovatively introduced as the induction factor. The extended GIOWA operator was configured with different parameters, and the maximization of the Dice coefficient was adopted as the optimal criterion, whereby a combination forecasting model for air cargo volume was established. Empirical results demonstrate that the proposed combination forecasting model based on the extended GIOWA operator is significantly superior to single forecasting models. An effective modeling approach and methodological reference are thus provided for air cargo volume forecasting, offering substantial practical value for scientific decision-making and refined management in the air logistics industry.