Combined Forecasting Model of Water Traffic Accidents Based on Gray-BP Neural Network

Abstract

In order to improve the prediction accuracy water traffic accidents, this paper combines the gray prediction model with the BP neural network model to establish a combined forecasting model of water traffic accidents based on gray-BP neural network. Based on the data of China’s national number of water traffic accidents, number of vessels, number of barges, number of crew, total transportation volume and water passenger volume, the result of prediction based on the combined prediction model prediction is compared with results of the gray prediction model and the BP neural network prediction model separately. The experiment proves that the combined forecasting model of water traffic accidents based on Gray-BP neural network has less error, higher prediction accuracy and better stability than the gray prediction model and BP neural network prediction model.