November 9, 2020
Source: AI Trends
The combination of IoT and machine learning growing at the same time is leading to a rise in the use of digital twins in the supply chain, as a digital replica that can be used for various purposes. The connection with the physical model and the corresponding virtual model is established by generating real time data using sensors.
The Digital Twin Consortium, launched in August as a program of the Object Management Group, is working on defining a taxonomy and standards and enabling technology including AI and simulation. Engineers are being attracted to the work. Founding members include Ansys, Dell, GE, Lendlease, Microsoft and Northrop Grumman.
“IoT and ML are the raw materials and the tools —the insight is in the repository where we model processes and create context. While this might be a database or a data lake, the most interesting example of this for me is the digital twin,” wrote Scott Lundstrom, an analyst focused on the intersection of AI, IoT and Supply Chains, on his blog, Supply Chain Futures.
The digital twin in the supply chain allows a comparison between current and historical data on performance, wherever a sensor is located. It could be a component such as a thermostat, an asset such as a truck or a machine, an employee such as a service technician, or a process, as in manufacturing. “Part of the capability of the digital twin is driven by this complexity of having models of models to describe complex assets, processes, and systems,” Lundstrom wrote.
In the supply chain, the digital twin model can encompass items packed in containers, moving through the physical world to distributors and customers. The model could inherit data from the process that created the product at one end of the chain, and inform a customer model at the other end.
“Supply chains and manufacturing assets are just the beginning. As this technology becomes better understood, and deployments become easier, use will grow into increasing complex spaces. There is already development of digital twins in life sciences in support of systems biology modeling complex organs like the human heart,” Lundstrom wrote. (See “Virtual Twins: Their Roles in Healthcare, Drug Discovery and Pandemic Response,” in BioITWorld.)
Ideally for the supply chain, characterized by many complex, multi-model use cases, the inclusive digital twin can have a view of the entire supply chain from the supplier’s supplier to the customer’s customer. An understanding of the status and history of assets and processes allows machine learning tools to be brought into the equation to execute simulations, optimizations, and predictive capabilities to the models, Lundstrom suggests.
“To realize the benefits of this tremendous opportunity we need standards, agreed upon taxonomies, and commercial development tools and platforms for this market to flourish,” he stated. “The supplier community is reacting to this opportunity, and many practitioners from the PLM [product lifecycle management], IoT, and analytics/data science market are beginning to focus on resolving some of these foundational standards.”
The large platform suppliers are moving forward with tools and platform as a service (PaaS) offerings to try to win share and develop “de facto” standards. Amazon Web Services (AWS), Google Cloud Platform (GCP), Predix Platform from GE, IBM and Microsoft are all building extensions to their existing IoT tools and platforms to add support for the creation of digital twins.
“By leveraging JSON-LD, a well-accepted and simple object framework, Microsoft is supporting an open standard from the beginning,” Lundstrom writes. “This is a key requirement as users begin to understand that digital twins require an open object-oriented approach to support the requirements for inheritance, and multiple instances in creating complex multitier models that are portable and support the use of widely available cloud platforms and AI frameworks.”
Are Supply Chain Digital Twins Just Another Fad?
Is the supply chain digital twin just another fad, asked a blog post on the site of River Logic, a supplier of prescriptive analytics technology for supply chain optimization using digital twins. In business since 2000 in Dallas, the company offers pre-built applications with knowledge of business planning and optimization.
Simulation and modeling software allows organizations to create realistic and verifiable supply chain digital twins of their supply chains. Data mining techniques along with inputs from Internet of Things (IoT) sensors allow real-time data to be fed into models. The models can monitor and determine what’s happening in the real world and plan the appropriate corrective action.
A Gartner study on IoT implementation in July 2018 showed that 13% of companies working with IoT projects already had digital twins, while another 62% were working toward their implementation. “It seems that digital planning twins are more than just a fad,” the River Logic post stated.
Engineers in the 1970s and ‘80s were using three-dimensional CAD models of complex engineering equipment to conduct virtual walkthroughs. As the CAD technology advanced, it became possible to represent physical stress, making it possible to conduct virtual stress testing. Today it is possible to construct “almost perfect” digital models of real equipment, such as aircraft, autonomous vehicles and drilling equipment, and by inputting real data, such as the static and dynamic loads experienced during aircraft takeoff, to measure performance.
“In this way, it’s possible to simulate the real world and bridge the gap between our physical and digital environment,” River Logic states. Several company experiences with digital twins are highlighted on the River Logic website.
Digital Twin of a Warehouse in Pacific-Asia Built by DHL Supply Chain
DHL Supply Chain built its first digital twin of a warehouse in Pacific-Asia for Tetra Pak, a multinational food packaging and processing company based in Switzerland. The digital twin is supplied with real-time data on a consistent basis from a physical warehouse in Singapore, which DHL developed to be integrated into the supply chain, according to an account in Supply Chain magazine.
“The joint implementation of such a digital solution to improve Tetra Pak’s warehousing and transport activities is an excellent example of the smart warehouses of the future,” stated Gillet Jerome, CEO, DHL Supply Chain Singapore, Malaysia, Philippines. “This enables agile, cost-effective and scalable supply chain operations.”
At the warehouse, the DHL Control Tower tracks incoming and outgoing goods to ensure all goods are stored in the correct way within 30 minutes of receipt. Incoming trucks are outfitted with IoT technology. A smart storage solution developed by Tetra Pak tracks and simulates the physical condition and individual stock levels in real-time, allowing non-stop coordination of operations. .
“We expect the partnership with DHL Supply Chain to further increase our productivity and maintain high standards in our supply chains,” commented Devraj Kumar, Director, Integrated Logistics, South Asia, East Asia & Oceania for Tetra Pak.
Digital Twins in Paris Will Protect Wind Turbine from North Sea Gales
GE engineers in Paris are partnering with Ansys, a global supplier of engineering simulation software, to build a digital twin of a wind turbine in the North Sea. One goal is to maximize output and minimize downtime by spotting problems before they lead to an unplanned outage. The predictive maintenance relies not only on physical sensors on the machines, but also virtual sensors put in places where physical sensors cannot be used, according to an account from GE News.
The virtual sensor has the ability to guess with fair precision a value such as temperature of pressure, by using other data from sensors and smart algorithms based on historical data or models.
For example, the GE engineers have developed a digital twin of the Haliade 150-6 wind turbine’s yaw motors, which enable the 6-megawatt turbine to rotate and position itself into the wind. Using virtual sensors, this digital twin simulates the temperature at various parts of the motors.
Using algorithms built on Predix, the GE software platform for the industrial internet, and a modeling approach developed by Ansys, the engineers can now estimate the motor temperature at any given moment. At the Foundry, they can also monitor how the motors perform under different strains over time. In the field, engineers are able to use an app with a dashboard connected to the twin, to monitor the motor’s temperature.
“For the simulation, thanks to the digital twin we only need to know the current to understand the temperature and optimize the use of the motor,” Sabot stated.
GE reports it has 1.2 million digital twins of jet engines, gas turbines and locomotives already working in the field.