In order to scientifically assess the operational efficiency of airport surface movements, a departure flight taxi time prediction method considering runway and taxiway system configurations was proposed. Firstly, the definition of surface taxi time was provided, and an analysis of the historical operational data at the airport was conducted to identify more accurate factors influencing surface taxi-out time. Based on the conclusions drawn from correlation analysis, a random forest prediction model for departure flight surface taxi time was constructed. The method was illustrated using actual operational data from Shijiazhuang Zhengding Airport and Shenzhen Bao'an Airport, analyzing the characteristics of surface movement time under different runway system configurations. The results indicate that: Apart from taxi distance, the correlation results of factors influencing surface taxi-out time at airports with different runway and taxiway system configurations are generally consistent. When considering the same factors, the prediction of surface taxi time is better for single-runway airports compared to dual-runway airports. Surface taxi time at single-runway airports approaches unimpeded taxi time, while at dual-runway airports, there is a significant difference from unimpeded taxi time. The research findings are of significant importance for enhancing airport surface operation efficiency and achieving energy savings and emissions reduction.