Requirements and design patterns for adaptive, autonomous, and context-aware digital twins in industry 4.0 digital factories
Paolo Bellavista, Nicola Bicocchi, Mattia Fogli, Carlo Giannelli, Marco Mamei, Marco Picone
Abstract
Digital factories are poised to achieve unseen levels of resiliency and flexibility, facing increasingly demanding requirements by customers and market conditions. Digital twins are one of the building blocks fueling this vision. They provide a software counterpart for industrial assets enabling control, simulation, analytics and “servitization” functionalities. To effectively fulfill their tasks, digital twins need to embed adaptive, autonomous, and context-awareness functionalities. In this work, we propose an organic vision of digital twin design and implementation with the goal of clearly identifying the primary steps towards this goal. First, we detail how current requirements for digital twins have to be enriched for supporting adaptivity, autonomy, and context-awareness. Second, we propose a set of reusable design patterns mostly popularized in the field of micro-services allowing engineers to meet these new demanding requirements while keeping complexity and management costs under control. Finally, we present our working prototype based on the identified design patterns and implemented with orchestrated micro-services, demonstrating the feasibility of our solution and quantifying its networking and computational overhead.