HOSPITAL MANAGEMENT: LOGISTICS MANAGEMENT SYSTEM, STOCK LEVEL MANAGEMENT, ABC ANALYSIS WITH DEMAND FORECASTING USING MACHINE LEARNING

Abstract

Inventory management involves purchasing, storing, utilizing, and selling a company’s stock, including components, raw materials, and finished goods. This capstone project applies technology-driven inventory management to improve hospital logistics. The system incorporates cloud-based databases for storing critical documents to optimize space, artificial intelligence (AI) for demand forecasting, and a user-friendly website to enhance accessibility. By adopting a modular, scalable, and secure architecture, the system improves efficiency, reduces human error, and enables real-time decision-making. To support the development and continuous improvement of hospital logistics, this study employed the Agile Scrum methodology. Each component of the system, including inventory management, listings, bidding and awarding, and demand forecasting, operates independently but is interconnected through RESTful APIs. The system was developed using a microservices architecture. MongoDB was selected for listings and inventory management due to its schema flexibility and strong performance with large data and user volumes, while Mongoose was used to facilitate database communication, reduce redundancy, and improve data retrieval speed. The predictive analytics module applies time series models to forecast demand, with a TensorFlow-based model trained on historical inventory data. This approach helps reduce shortages and overstocking while ensuring optimal stock levels. The deployment of the Hospital Logistics System led to significant improvements in tracking, efficiency audits, and inventory infrastructure management. Real-time inventory updates were integrated to maintain consistent stock levels, while the system also monitored items nearing expiration and facilitated their location within the warehouse. The AI-powered demand forecasting algorithm generated highly accurate predictions, enabling staff to make real-time replenishment decisions based on forecasted demand. The findings demonstrate that automating inventory processes significantly improves productivity while reducing human error. Integrating AI for demand forecasting provides a proactive approach to inventory management, ensuring that stock levels align effectively with demand patterns.