Abstract:
The increasing adoption of Point of Sale (POS) systems in Small and Medium-sized
Enterprises (SMEs) has underscored the need for advanced functionalities such as
real-time inventory management and customizable role-based access. Traditional POS
systems often lack the flexibility to meet diverse SME operational requirements,
limiting efficiency and personalized user control. This project addresses these
limitations by integrating real-time inventory tracking and adaptive role-based access
control to enhance operational efficiency and system usability.
The system follows a structured development approach, incorporating a literature
review, requirement gathering through interviews and surveys, and a microservices
based architecture. The back-end is built with Spring Boot, while the front-end
leverages React for an interactive user experience. Additionally, machine learning
models, including Decision Trees and Random Forest, enable predictive sales
forecasting and dynamic inventory optimization, allowing SMEs to anticipate demand
and minimize stock discrepancies.