In October 2018, Microsoft launched in preview Azure Digital Twins, a platform that enables the creation of knowledge graphs — models of a knowledge domain created by subject-matter experts with the help of AI algorithms — based on digital models of entire environments. After two years of development, Microsoft today announced that Azure Digital Twins is generally available, and that the Digital Twins Consortium, a collaborative effort cofounded by Microsoft in May 2020 to drive consistency in digital twin technology, now has over 170 members spanning companies, government agencies, and academia.
Digital twin approaches to simulation have gained currency in a range of different market segments. For instance, London-based SenSat helps clients in construction, mining, energy, and other industries create models of locations for projects they’re working on, translating the real world into a version that can be understood by machines. GE offers technology that allows companies to model digital twins of actual machines and closely track performance. And Oracle has services that rely on virtual representations of objects, equipment, and work environments.
Azure Digital Twins enables users to model environments such as buildings, factories, farms, energy networks, railways, stadiums, and cities by connecting assets like internet of things devices and existing business systems. The platform’s event component supports the creation of dynamic business logic and data processing; Azure Digital Twins integrates with Azure data, analytics, and AI services, helping users to track the past and then attempt to predict the future.
In Azure Digital Twins, customers define the digital entities that represent the people, places, and things in their physical environment using custom twin types called models. Models are defined in a language called Digital Twins Definition Language (DTDL), and they describe twins in terms of their state properties, telemetry events, commands, components, and relationships. For their parts, models define semantic relationships between entities so that companies can connect twins into a knowledge graph that reflects their interactions.
Digital models in Azure Digital Twins are live, up-to-date representations of the real world. Using the relationships in DTDL models, customers can connect twins into a live graph representing their environment. Users can view a visualization of their Azure Digital Twins graph through the Azure Digital Twins explorer and leverage the platform’s event system to keep the graph current with data processing and business logic. Customers can connect external compute resources to drive this data processing, extract insights from the live execution environment using Azure Digital Twins’ query API, or route data downstream to Azure services for analytics or storage via event routes.
According to Microsoft, Korea-based Doosan Heavy Industries & Construction worked with Azure Digital Twins and Bentley Systems to develop a digital twin of its wind farms, which allowed operators to remotely monitor equipment performance and predict energy generation based on weather conditions. Meanwhile, Johnson Controls is collaborating with Microsoft to develop new digital twin-based architectural design tools. Ansys now offers native integration with Azure Digital Twins through its Ansys Twin Builder tool. And Brookfield Properties partnered with Willow to create a digital replica of their One Manhattan West (OMW) property using Willow’s Azure Digital Twins-powered product.
“According to the IoT Signals report, the vast majority of companies with a digital twin strategy see it as an integral part of their IoT solution. Yet the reality is that modeling entire environments can be complicated,” Azure IoT corporate vice president Sam George wrote in a blog post. “Azure Digital Twins is an industry first. It breaks down silos within intelligent environments by fusing data from previously disparate devices and business systems.”
The public launch of Azure Digital Twins follows the unveiling of Microsoft’s Project Bonsai, an AI development platform for industrial systems. Project Bonsai is a “machine teaching” service that combines machine learning, calibration, and optimization to bring autonomy to the control systems at the heart of robotic arms, bulldozer blades, forklifts, underground drills, rescue vehicles, wind and solar farms, and more. Control systems form a core component of machinery across sectors like manufacturing, chemical processing, construction, energy, and mining, helping manage everything from electrical substations and HVAC installations to fleets of factory floor robots.
According to Markets and Markets, digital twins services will be worth $48.2 billion combined by 2026, up from $3.1 billion in 2020. The firm attributes the rise to the changing face of maintenance, an uptick in interest among automotive and transportation sectors, and the growing adoption of digital twin solutions in pharmaceutical and health care businesses to cope with the COVID-19 pandemic.