Implementation Augmented Intelligence on Drug Inventory Management Forecasting

The administration of medications at many hospitals is not governed by an inventory forecasting system, which is common in the management of many hospitals. This might either result in an excessive amount of pharmaceuticals being available in the hospital or a serious shortage of them. Because of this, it is essential for the hospital to have inventory forecasting for the management of drugs; this will assist the hospital in determining when the medicine will run out of supply. As a result of this, it is essential for the hospital to have inventory forecasting for the management of drugs. The author of this study builds an augmented intelligence with the help of a stacked LSTM model. Augmented intelligence is a kind of intelligence that supports stakeholders in getting a broad grasp of inventory projections for the foreseeable future. Using event-driven architecture, the systems are developed independently so that they will not interfere with the workings of the current hospital administration system. We test the model on 19 different drug stocks using a total of 1000 days of data for each medication, and the results reveal that stacked LSTM is able to correctly predict with average values for the R2 coefficient, RMSE, MSE, and MAE being 0.9972, 1.95, 3.81, and 1.26 respectively. The results also show that the proposed system can work.