Large trucks are an important component of the transportation system in open-pit coal mines. The effective utilization of trucks not only directly affects the progress of the project but also has a significant impact on the economy of open-pit mining enterprises. To effectively solve the truck failure prediction problem in open-pit mines, a BP neural network model based on an improved gray wolf optimizer is proposed to predict the number of truck failures and the duration of truck failures in open-pit mines. This method first addresses the shortcomings of the traditional gray wolf optimizer’s weak convergence performance and local escape extreme value performance by integrating Circle chaotic mapping, nonlinear update mechanism, and population update method based on linear interpolation to propose an improved gray wolf optimizer (IGWO) and verifies the effectiveness of the IGWO algorithm through comparison with 6 benchmark functions and 6 algorithms. Secondly, using the better optimization performance of the IGWO algorithm to find the optimal weights and thresholds in the BP neural network model, an IGWO-based BP neural network model (IGWO-BPNN) is proposed. Finally, taking the truck failure data of the Baorixile open-pit coal mine as an example, the effective prediction was performed using the IGWO-BPNN model. Experimental results demonstrate that, under identical experimental conditions, the IGWO-BPNN algorithm exhibits superior predictive performance when compared to traditional BP neural network models and BP neural network models based on traditional GWO. The resulting predictive outcomes not only facilitate the development of scientifically-informed preventive maintenance plans for open-pit mining enterprises but also provide a robust foundation for data-driven decision-making in the construction of intelligent open-pit mines.