Abstract: The introduction of autonomous mining trucks (AMTs) has the potential to effectively mitigate existing issues related to low efficiency, high labor costs, and numerous safety hazards in open-pit mining operations. However, due to limitation of autonomous driving technology, at each intersection the passage of only one AMT is permited at a time to ensure safety requirements. Consequently, queues of waiting AMTs at intersections are likely to form, leading to heightened emissions of greenhouse gases. Additionally, the dynamically changing operational conditions within the mining area necessitate real-time scheduling for AMTs. This study centers on the scheduling of AMTs in open-pit mining operations. A real-time data-driven dynamic scheduling simulation platform was devised, accurately representing queues at loading and unloading zones as well as intersections, while adhering to a first-come-first-served principle. With the aims of minimizing greenhouse gas emissions and maximizing production output, an AMT scheduling model was constructed in time-space network. Furthermore, a heuristic priority-based scheduling algorithm was formulated and evaluated. This algorithm achieved rapid responsiveness at the millisecond level, suitable for real-time scheduling requirements. Leveraging actual road network and production data from an open-pit mine in northeastern China, scenario simulations and sensitivity analysis were conducted employing different strategies. Results show that in a bustling open-pit mine, scheduling strategies accounting for intersection congestion not only curtail greenhouse gas emissions, but also enhance production efficiency compared to strategies overlooking such congestion factors.