Predictive Sales and Inventory System with Customer Segmentation for Enhance Customer Relationship Management

Abstract

This study presents the design and development of a Predictive Sales and Inventory System with Customer Segmentation for Enhanced Customer Relationship Management. The system addresses operational inefficiencies in small retail businesses by integrating predictive analytics and clustering techniques. Linear regression was applied for sales forecasting, while K-means clustering was used for customer segmentation to support data-driven decision-making and personalized customer engagement. The platform automates sales transactions, inventory monitoring, and membership management, and supports RFID-based cashless payments. Development followed the Agile Scrum methodology, with modules including product management, transaction processing, analytics, reporting, and backup management. User Acceptance Testing based on the Technology Acceptance Model (TAM) yielded an overall weighted mean score of 4.68, indicating strong user acceptance, with one establishment expressing intent to adopt the system. System evaluation based on ISO/IEC 25010 quality standards produced an overall weighted mean score of 4.18, reflecting satisfactory performance in functionality, reliability, usability, and security. Results demonstrate that the system provides an efficient, browser-based, data-driven POS solution that improves transaction speed, operational accuracy, and business insight generation for Spot777 Coffee.