Main Article Content
Abstract
Accurate sales forecasting is crucial for businesses to optimize inventory, manage resources efficiently, and improve decision-making. This thesis presents a comprehensive sales forecasting dashboard that integrates time series forecasting models such as ARIMA, SARIMA, and Prophet. The dashboard, developed using Streamlit, provides an interactive interface for users to upload sales data, visualize trends, detect seasonality, and generate future sales predictions. The proposed system enhances business intelligence by providing actionable insights and facilitating informed strategic planning. This study demonstrates the potential of automated forecasting tools in transforming retail analytics. By improving the accuracy of sales predictions, businesses can achieve better financial planning and strategic management, ultimately enhancing overall efficiency and profitability. The research also highlights opportunities for future advancements, such as incorporating external economic factors and exploring deep learning techniques to further refine forecasting accuracy. This thesis presents the development of a Sales Forecasting Dashboard using Streamlit, leveraging machine learning models such as ARIMA, SARIMA, and Prophet. The dashboard provides businesses with real-time insights into sales trends, seasonality patterns, and accurate sales predictions, enhancing decision-making processes. Historical sales data is analyzed to detect trends and seasonality, and multiple forecasting models are implemented to ensure accurate future sales predictions. The dashboard is designed to be interactive and user-friendly, allowing for seamless data uploads and dynamic visualization