Iot and Machine Learning Based Smart Pharmacy Inventory Management For Drug Demand Prediction

Abstract

The management of inventory in a pharmacy is critical in ensuring that medicines are available on a continuous basis while minimizing overstocking, waste and supply chain inefficiencies. The conventional method of managing inventory in a pharmacy includes manual monitoring and use of historical data to guide ordering of medicines. This method is inadequate as it does not account for the dynamic nature of consumption of medicines by patients. The incorporation of Internet of Things (IoT) technology and machine learning (ML) into the management of inventory in a pharmacy can result into an intelligent system for real time monitoring and accurate prediction of demand for drugs. This study proposes an IoT and ML-based framework for management of inventory in a smart pharmacy. The framework comprises of two layers, namely, the IoT layer which comprises of sensing devices that are connected and which are used to automatically capture data that are related to quantity of medicines, storage conditions of medicines, expiry dates of medicines and patterns of consumption of medicines. The data that are automatically captured are processed and analyzed by using ML algorithms in order to determine trends in demand and to predict future requirements of medicines. Several predictive models can be evaluated on the basis of their performance in forecasting by means of statistical measures that evaluate performance such as Mean Absolute Error (MAE), Root Mean Square Error (RMSE) and Mean Absolute Percentage Error (MAPE). The intelligent framework proposed in this study automatically makes decisions regarding management of inventory in a pharmacy by means of stock replenishment recommendations that are timely and which can result into prevention of several occurances such as shortages of medicines and losses due to expiration of medicines. The real time sensing capability in conjunction with predictive analytics leads to efficient management of resources in a pharmacy and also aids in establishment of a smart and sustainable health care supply chain management system.