A Review on Modern Trends in Digital Twin Technology in Retail Stores

Abstract

Digital twin (DT) technology has emerged as a transformative solution for managing the operational complexity of modern retail grocery stores. This systematic review examines seven contemporary DT methodologies applied to retail inventory management, store organization, and supply chain resilience, comparing their approaches, limitations, and practical contributions. The reviewed studies span a diverse range of technical implementations, from mathematical optimization using Wilson's Economic Order Quantity model to AI-driven computer vision pipelines combining YOLOv8 detection with Vision Transformer classification for real-time shelf mapping. Hardware-based approaches are also examined, including autonomous RFID robots achieving 99.06% inventory accuracy and portable Bayesian-filter localization systems integrated with SLAM-based 3D reconstruction. Complementing these technical studies, conceptual frameworks and qualitative case studies address broader challenges such as supermarket layout management, GDPR-compliant DT architecture design, and zero-waste sales and operations planning in grocery retail. The review identifies several persistent research gaps that constrain wider adoption: limited end-to-end value chain integration, inadequate handling of perishable goods and expiry tracking, weak privacy governance frameworks, real-time synchronization latency, and the absence of standardized performance benchmarks across studies. The COVID-19 pandemic is highlighted as a catalyst that exposed the inadequacy of traditional planning systems and accelerated demand for simulation-based resilience strategies. The findings provide both technical blueprints and strategic frameworks for practitioners and researchers, establishing a foundation for scalable, interoperable, and privacy-conscious digital twin deployment across multi-store retail environments.