DigiPyramid: A Multiresolution Digital Twin Framework for Reliable Logistics Management-Conceptualization, Implementation, and Intelligentization

Abstract

This article presents DigiPyramid, a multiresolution Digital Twin framework that enhances reliable logistics management. DigiPyramid leverages Robotic Process Automation for low-cost, flexible data aggregation across heterogeneous systems like ERP, WMS, and TMS. The framework employs Bayesian Networks to infer and mitigate ambiguities such as ‘‘breakpoints,’’ ‘‘black boxes,’’ and ‘‘gray boxes,’’ ensuring data reliability. Additionally, it utilizes Analytic Hierarchy Process to tailor decision-support indicators for users at different organizational levels — managers, planners, schedulers, and operators — enabling context-specific resolutions. A case study of a Chinese national logistics enterprise demonstrates DigiPyramid’s effectiveness in improving visibility, adaptability, and coordination, particularly during disruptions like COVID-19. The integration of AI technologies, including Computer Vision and Bayesian Networks, further enhances real-time monitoring and predictive capabilities. By aligning digital transformation with sustainability goals, DigiPyramid offers a scalable solution for resilient and efficient logistics operations, contributing to both theoretical advancements and practical applications in logistics process management.