Application of medical supply inventory model based on deep learning and big data

The existing management structure of medical supply inventory (MSI) is not sufficiently effective, and it is incompetent to solve the problems of medical supply stock control in public security emergencies. Therefore, deep learning and big data technology are employed in this work to optimize the stock control structure and enhance management efficiency, so that the optimized management structure can play an excellent role in the material supply of emergencies. After browsing copious literature, the economic ordering models with infinite/limited supply rate and without shortage are innovatively constructed to realize efficient management of emergency supplies inventory. Besides, the optimized fixed-point and quantitative ordering method of safety stock is employed to construct the MSI models for scarce emergency supplies and the time-sensitive emergency supplies, respectively. Then, an earthquake-related emergency is taken as a case and data source to evaluate the solution results of the emergency MSI model. Moreover, the stacked auto-encoders (SAE) algorithm is used to build the demand prediction model for MSI. Finally, a simulation experiment compares the SAE-based demand prediction model for MSI with a back propagation neural network (BPNN) model and radial basis function network (RBFN) model to verify the model’s performance. The experimental results demonstrate that after 150 times of training, the error between the predicted value and the actual value of each model is within 30, and the prediction accuracy is significantly improved. After 170 times of network training, the mean absolute error (MAE) values of BPNN model and RBFN model are 31.98 and 73.73, respectively. In contrast, the MAE value of the SAE-based model is 21.32, which is superior to the other two models. Evidently, the management structure of MSI is optimized by dividing the emergency MSI into three MSI models for the critical emergency supplies, scarce emergency supplies, and the time-sensitive emergency supplies. The research outcome can provide essential logistical support for dealing with public security emergencies.