Management of Mechanical Transport Equipment and Analysis of Return Rate by Using BP Neural Network

Based on the single mechanical transport equipment report of MCC, by using BP neural network with a single hidden layer from MATLAB, the company's return rate in next month is forecast. Propose a scheme that improve the input layer, the hidden layer and its node to ensure the high efficiency of training and the speed of convergence rate. After using the improved model, with only 150 training, the expected error reach 0.0005. Subsequently, further screening the selected data for equipment that affects the rate of return, combining with the fluctuant return rate of transportation equipment to set a screening standard for high utilization rate and high risk rate. Analyzing the problems arising from the emergence of dual high equipment to put forward the equipment management requires that replace ‘repair’ with ‘inspection’ and replace ‘change’ with ‘repair’.