Abstract: The purpose is to develop and apply a machine learning-based approach for the prediction and scheduling optimization of open-pit mining truck speeds. Open-pit mining is a significant method for coal mining, making it crucial to enhance the transportation efficiency of mining trucks. Using machine learning to achieve accurate prediction of vehicle speed, in order to improve production efficiency, reduce costs, and enhance work safety. The methodology encompasses data cleansing, curvature and gradient calculation, and the construction of machine learning models. Firstly, the data was cleaned to eliminate noisy data, and the time and device information were converted. Next, use latitude and longitude data to calculate curvature radius and slope, in order to more accurately describe road conditions. Finally, machine learning algorithms such as random forest and XGBoost are used to predict vehicle speed based on onboard data and weather sensor data. The experimental results demonstrate that machine learning-based models can predict open-pit mining truck speeds highly accurately. Among these models, the Random Forest-based model exhibited lower mean squared error and a higher coefficient of determination, outperforming the XGBoost-based model. The predictive performance of these models provides robust support for production and scheduling. The conclusion is that machine learning holds substantial potential in open-pit mining truck speed prediction and scheduling. This technology is poised to enhance the precision of loading and unloading tasks, reduce resource wastage and waiting times, alleviate traffic congestion, and improve production efficiency. Additionally, it aids in early prediction of hazardous situations such as speeding, thereby enhancing work safety. Machine learning also supports real-time decision-making to adapt to constantly changing circumstances. While the study primarily focused on the prediction of open-pit mining truck speeds, machine learning techniques have extensive applications in logistics, mining, and other domains. This research serves as a strong example for exploring the application of machine learning in the industrial and mining sectors, offering insights for future research and innovation.