The company productivity enhancement process starts with an overview of the product and its processes from a bird’s eye view. External business consultants and in-house managers try to take stock of the situation and emphasize the critical zones that block the successful development of a business. The effectiveness of such a process is low due to the lack of real-time data bounded with the ingeniousness of visualization and analyzation processes. In some cases, the product and processes are incredibly complicated, so it is hard to predict the consequences of the actions. Additional complications come from the necessity for businesses to be agile and make decisions fast.
One of the most powerful solutions came in 2002 at a Society of Manufacturing Engineers conference in Troy. Dr. Michael Grieves had introduced the Digital twin concept. The idea unites Data Science, Artificial Intelligent (AI), Machine Learning (ML), and the Internet of Things (IoT) in one solution.
What are digital twins and digital threads?
The digital twin concept refers to the digitalization of the Product Lifecycle Management process and paves the way for fascinating opportunities for improvements in the manufacturing and process industries. Creating a digital twin means to form the virtual replica of the product, mimicking the interfaces, settings, qualities, and dynamics. All the data is gathered in the cloud-based systems with individual sensors that are integrated into the real products. The digital twin is a 3D model of the physical good with the same metrics as a real one. The process is constant, and data can be varied and differ in size, depending on the capacity of the Internet of Things (IoT) device and goals laid out before the model. The platform provides manufacturers with an output rendered virtually. Research and development departments(R&D) use digital twin of the real product to recognize weak zones and opportunities that can be used. Its virtual implementation helps in the development of new product designs, test new solutions with real-time feedback, and evaluate the effectiveness of the solution.
Companies expand the usage of the digital twin’s concept by starting with the improvement of business processes and workflows. First of all, they gather all the data about the business in one place: manufacturing operations, supply chain, data warehouse, etc. A digital thread helps better understand the data-flows framework that exists across these systems. R&D departments begin by analyzing these information flows and search for ways to help businesses leverage their operations: predicting maintenance issues, avoiding downtime, improving customer service, crowd management, and security. Thereby, the digital twin as a dynamic model is significantly useful for the company’s maintenance of its connected machines and equipment units.
The digital twin approach stayed on high tide for the last few years and was included in Gartner’s Top 10 Strategic Technology Trends for 2017 and 2018. The new wave of IoT will allow this concept to remain intact during the next few years. According to Gartner’s research, 24% of organizations that either implemented the Internet of Things (IoT) solutions in production or IoT projects in progress already use digital twins. We need to mention that another 42% plan to use twins within the next three years.
Digital twin architecture
The internal or external R&D team creates digital twins. Minding the goals of the business, the digital twin concept can be applied directly to a product, a process, or an entire system. However, the architectural approach still consists of the following 4 steps:
- Model creation: The first step of model creation starts with the product analysis and definition of all the characteristics that the product has. After the analysts check the procedure connected with the process of creating, storing, delivering to the end-user, and exploitation, they integrate sensors with the physical products and cover the process with different IoT devices. These elements collect information about sizes, states, colors, and various physical performance criteria. After the data is transmitted, typically to the cloud-based platform, it provides the digital twin with a wide range of data that is continually updating. The model is created based on the input from the sensors. To stabilize the interaction between the model and the real product, a communication protocol needs to be established. The main requirement is that the model needs to have seamless, real-time, bidirectional connectivity between the physical process and the digital twin.
- Experiment: At this moment, a business would already have the digital twin that is ready for the experiments. R&D specialists have a list of hypotheses and improvements that they want to check. They start by performing different experimental procedures. All the results of the testing are gathered in the data lake and are subsequently ready to be processed and analyzed. The data can be processed either on the premises or in the cloud.
- Analyze: The next, and one of the most essential steps, is analysis. Data scientists generate insights, recommendations, and guide decision making. They give the business a visual presentation of the process and highlight the differences in the performance of the digital twin model and the physical world analog in one or more dimensions. Experts indicate areas that potentially need investigation and alteration.
- Act: At this stage, a company should understand the steps that need to be taken to improve the product or the process. They have the digital twin with the necessary changes and improvements. C-Level has enough metrics and prediction information to approve the solution. If the improvement is approved, the company starts to implement it. Otherwise, R&D teams continue the investigation processes.
The Digital concept is nourished by the Deming Cycle that is used by the leaders of the Market. It is the basis for constant product and process improvements.
Digital Twins solution providers
Word IT leaders have started to create solutions that simplify the implementation of Digital twins. Here are two of the most compelling examples:
- Azure Digital Twins. Microsoft believes that IoT is the key to business transformation, and have indicated that 88 % of companies credit IoT as critical to their success. That’s why they actively develop Azure Digital Twins as a system that integrates Cloud, IoT, Edge Computing, Artificial Intelligence, and Mixed reality. They offer the possibility to easily integrate Azure Virtual Machines, Managed Disks, Azure Cosmos DB, and HockeyApp to extract raw data from experiments. Azure Databricks and HDInsight provide the possibility to analyze the results by creating spatial intelligence graphs. The solutions are robust and scalable. By understanding the value of the IoT and ways it can empower businesses, Microsoft has become one of the leading companies in the digital twin’s solutions for small businesses and enterprises.
- IBM. There is no need to introduce this multinational information technology company. International Business Machines Corporation was born in the USA and has since spread to over 170 countries all over the world. IBM announced new lab services for Maximo. They want to bring Augmented Reality (AR) into asset management. They have a partnership with DAQRI, which is a leading vendor of augmented reality for industrial use. The IBM lab introduces visual and voice features for workforces. They use Natural Language Processing, ML, and AI sets. The implementation is empowered by DAQRI Smart Helmet™ which enables a business to see their assets in a new dimension and obtain instant access to critical data.
Digital Twins cases:
Let me bring to the table a few examples of how digital twins help a real business achieve success
- Stara and the improvement of the farming business: Stara integrates their equipment with IoT sensors that extract information about the performance, location, and state of their agricultural machinery. With real-time information, Stara can proactively prevent equipment malfunctions and breakdowns. On another side, Stara has been making further headway by improving the agriculture industry as a whole. After they upgraded their own business, Stara began to sell their solutions to the other farmers. The results are incredible. Those companies that adopted their solutions have reduced seed use by 21% and fertilizer use by 19%, thanks to Stara’s guidance.
- Kaeser and the changes made to their business model: Implementation of the digital twins has allowed Kaeser to switch from product to service selling. They changed their payment plan from a fixed rate to a fee that is based on air consumption. Kaeser teams installed their equipment with monitoring systems that provide real-time information about its state and performance. The company has cut commodity costs by 30% and onboarded 50% of the major vendors using digital twins.
- GE Health Innovation Village: GE is another world-leading company that doesn’t require any additional introduction. General Electric put additional attention on healthcare technology. They opened the Health Innovation Village in Helsinki at the end of 2014. This incubator helps healthcare startups create innovative products and services based on the usage of wireless technologies, sensors, as well as apps and cloud services. The startups are covering technology innovations that feed into many aspects of healthcare, such as comfort for premature babies, monitoring, maintaining fitness, and other challenges. Implementation of the GE digital twins’ solutions has enabled heartbeat, blood pressure, respiration, and other information to be constantly streamed into the cloud where software can analyze it, alert doctors to anomalies and looming crises, and effectively create models that are used for innovations.
The future of digital twin
Thousands of companies across the world are staying ahead of the digital disruption by implementing the digital twin approach in their business. The usage of the digital twin concept is powered by the constant development of IoT and Data Science that give access to real-time measurement results: the computation power of big data, the versatility of the analytics technologies, flexible storage, greater possibilities within the aggregation area, and integration with canonical data. According to Gartner’s prediction, by 2021, half of the large industrial companies will use digital twins. As a result of this development, those organizations will be gaining a 10% effectiveness improvement.
Certainly very impressive, so let’s see where it leads.