Research on the construction of early warning model of enterprise budget execution based on BP neural network

This paper constructs a set of budget execution early warning model based on BP neural network. By extracting key variables such as budget amount, actual expenditure, deviation rate and execution period as inputs, multi-dimensional training samples are established, and the back propagation algorithm is combined with network training and parameter optimization. The experiments are validated using the actual financial data of a manufacturing enterprise, and the results show that the model has high accuracy and stability in budget anomaly identification, which is significantly better than the traditional rule-based early warning methods. The study shows that the BP neural network can be effectively applied to the dynamic monitoring and early warning of budget execution, providing prospective decision support for enterprises.