 This paper proposes a novel approach for assessing safety risks in underground coal mines. It uses a Particle Swarm Optimization Backpropagation Neural Network, PSOB PNN, to analyze historical data from the Shounen coal mine. The PSOB PNN model was found to be highly accurate, with mean square error, MSc, mean absolute percentage error, MAPE, and root mean square error, RMSc, values of 2.0 times 10 to the power of negative 4, 4.3%, and 0.92, respectively. This model can be used to accurately assess and predict safety risks in underground coal mines. This article was authored by Dorcas Mwadi Malumba, Jian Kong Lu, Jian Hao, and others.