Inventory Prediction Using ABC Analysis and Decision Tree Algorithm

This study proposes a prediction model to address inefficiencies in conventional inventory management within the eyewear retail industry, which often result in stockouts, overstocking, and excessive stock visibility. To mitigate these challenges, the project introduces a data-driven solution integrating ABC analysis and a decision tree algorithm to optimize inventory control. Sales data from a local optometry shop was analyzed. ABC analysis was applied to categorize inventory items based on their popularity, enabling more effective stock prioritization. The study follows the CRISP-DM methodology: business understanding, data understanding, data preparation, modelling, evaluation, and deployment. The predictive model achieved 92% accuracy. The final output was presented in a comprehensive dashboard that visualizes the ABC analysis results, and the number of predicted items sold for each month. This solution is ready for immediate implementation and offers scalability for future enhancements to the supply chain. By improving inventory management, the project contributes to Sustainable Development Goal (SDG) 8, which promotes sustained, inclusive, and sustainable economic growth.