Digital twin, the digital representation of a real object or system
Digital double (digital twin)
Research and Advisory company Gartner gives a very short definition:
Digital double is a digital representation of a real object or system.1
An extended definition might be:
Digital double (digital twin) is a software analogue of the physical device modeling the internal processes, specifications and behavior of the real object in terms of interference and the environment. An important feature of the digital double is that in order for his input effects uses information from sensors are real devices working in parallel. Operation is possible in both online and offline. Further, it is possible to conduct comparisons of information virtual sensors digital double with sensors of the real device, detection of anomalies and their causes.2
Installation of sensors on the real device is in the process of introduction at the enterprise of technologies for the industrial Internet of things (IIoT).
Without the creation of digital doubles of products it is impossible the introduction of modern technology PLM (Product Lifecycle Management, management product lifecycle). IIoT and PLM — the inherent attributes of the “smart factory” (Smart Factory). Its characteristic feature is the formation and use of a digital model of material flow, i.e., the digital double is no longer a separate product, and production system. All of the mentioned technologies — approaches to implementing the concept of the Fourth industrial revolution (Industry 4.0). If in the traditional industries the achievement of the desired characteristics of the product is conducted through numerous field tests, the Industry 4.0 seeks to carry out multiple tests with a digital double, and full-scale tests pass the first time.3
Digital double products includes:
- Geometric and structural model of the object.
- Set the calculation data of parts and products in General.
- A mathematical model describing all occurring in the product of physical processes.
- Information about the technological processes of manufacture and Assembly of the individual components and products in General.
- Management system product lifecycle.
Digital double is used at all stages of the product lifecycle, including design, manufacture, operation and disposal.
At the stage of conceptual design: created versions of the computer models developed products to assess and select possible solutions.
At the stage of technical design: selected in the previous step the option is being finalized and refined with the use of the models of elements. The resulting model allows to take into account and optimize the interaction of all the elements subject to operation and environmental influences, it is already possible to call a digital double of the developed product.
At the production stage: the developed model helps to determine the required manufacturing tolerances to achieve the required performance and ensure trouble-free operation of the product during the entire lifetime, and allows you to quickly identify the causes of faults in the testing process.
During the operational phase: model of the digital double may be modified and used to implement the feedback to make adjustments to the design and production of products, diagnostics and prediction of malfunctions, increase efficiency, to identify new needs of consumers.
Classification of twins products:
Digital twins-prototypes (Digital Twin Prototype, DTP). DTP-double contains the information necessary to describe and create physical versions of copies of the product. This information includes the geometrical and structural models, technical requirements and specifications; the cost model, the calculated (design) and technological models of the product. DTP-double can be considered a conditional permanent virtual product model.
Digital copies are the copies (Digital Twin Instance, DTI). DTI-doubles articles describe a specific physical instance of the product with which the double remains connected throughout the life span. Twins of this type are based on DTP-double and additionally contain the manufacturing and operational models that include the history of the manufacture of the product, applicability of materials and components, as well as the statistics of failures, repairs, replacement of components and assemblies, etc. Thus, DTI is a double product undergoes changes in accordance with changes in the physical copy when it is operating.
Aggregated counterparts (Digital Twin Aggregate DTA). DTA-twins products are defined as information management system physical instance of a family of products, which has access to all of their digital counterparts.4
The classification doubles the production system:
- Digital double the entire production system (PS)
- Digital production line double
- Digital DoppelgangeR of a specific asset in the production line
Digital double PS includes:
- Engineering model PS containing the digital description of the resources of the enterprise, the structure of production assets, means of technological equipment, the range and technology of products, the system of collecting information about the current condition of the equipment.
- The operational model of SS, which is the digital platform for describing the logistic structure of an enterprise, formation of schedules for manufacture of products, interplant and external cooperation, including the regulations of technical maintenance and repair of equipment. Mathematical description is also subject to the dynamics of internal material flows on the basis of digitalization which to generate optimal production schedules of works.
The most difficult for practical implementation is the operational model of the digital double PS, which, in particular, shall have the following functions:
- To carry out the necessary calculations for management decision making.
- To display real-time production processes in the production system.
- Experiments “what-if” mathematical modeling of production processes.
Another important task of the operational model of the digital double of the production system is to minimize possible failures of process equipment due to timely maintenance and repair (MRO). One of the best types is a predictive maintenance (PdM). It allows you to make the repair not according to a preconceived plan, and when you need it.
Meant not the elimination and prevention of equipment failures through interactive assessment of its technical condition on the totality of the data coming from sensors, and to determine the optimal timing of repair work. Digital double can serve as one of the tools of predictive maintenance, which allows you to simulate various options for full and partial waivers, the operation of the devices taking into account their modes of operation, environmental influences and varying degrees of wear.6 Digital counterparts allow maintenance crews to be well informed and to arrive at the place of work with all necessary spare parts, tools and instructions required for maintenance.
At the level of the operational model of a digital double of the PS functions of the MRO are considered as additional operations that are optimized in the operational production plan so that they minimize impact on the rate of passage of the workpieces through the production assets of the enterprise. This task, assume the MES-system (Manufacturing Execution System — software designed for operational production scheduling.)
It is important that digital double SS is maintained up to date through the implementation of its operational assets, taking into account the current status of manufactured products. To solve this problem are IIoT technologies. It can provide a connection of sensors, actuators and other equipment data collection with existing systems of production management and c performance model of the digital double PS.
Thus, the digital double — the fundamental notion of “smart factory” (Smart Factory) that you want to associate with the product (in this case used the term “double digital products”), and the process of manufacturing the products — in this case, you should use the term “double digital production system”. Shared at every stage of the product life cycle (PLM), these digital twins should be functionally linked and to provide operational characteristics of the designed and manufactured products in accordance with its purpose.4
The implementation of digital doubles — quite a serious project, which may require large investments and preliminary estimates of payback. However, according to research company Deloitte, the technology is rapidly expanding to various industries including aerospace, retail, healthcare and other sectors. Digital twins accelerate the development of products and processes, optimize the execution of the work and help in preventive maintenance.7 According to the report, the global market of technologies of digital doppelgangers in 2023 will reach $16 billion, while the turnover of the market of technologies that form the basis for this progress (in particular, IoT and machine learning), by 2020 is expected to double.
Areas of the most intensive growth in the use of digital doubles, is likely to be resource-intensive industries, such as manufacturing, oil and gas industry, aerospace and the automotive industry. However, these technologies also have application prospects in the retail trade, healthcare and in designing “smart” cities.
Given that digital doubles are supported by IT giants including IBM and SAP, companies today should be paid to these technologies the most attention. “Digital twins can significantly enhance the ability of enterprises to make proactive decisions based on data, to increase their effectiveness and get rid of potential problems, says the Deloitte report. They can also provide the opportunity for safe and economical way to work out scenarios of “what if”, that is essentially to experiment with the future”.8
According to Gartner, as of the beginning of 2019, 13% of organizations implementing IoT projects are already using digital doubles, while 62% are either in the process of creating digital doubles, or plan to do so.
“The results — especially compared with previous surveys — show that digital counterparts are gradually coming into mass use, said Benoit Lerwa (Benoit Lheureux), Vice President of Gartner research. We predicted that by 2022, more than two thirds of companies who have implemented IoT, deployed at least one digital double to manufacture. We can achieve this figure within a year”.
There are two reasons for this sudden demand for digital counterparts. First, they bring tangible business value and become indispensable for the IoT and digital strategies. Secondly, according to Gartner, the rapid growth of implementation due, in particular, active marketing and organization of training by suppliers of these technologies.9
In the system IT-Enterprise has a product “Managing product lifecycle (PLM), providing a multivariate design of new products, configuration management and performance, and managing a digital layout for each serial number or batch of products throughout the life cycle.
The product “Industrial Internet of things (IIoT)” report provides data collection equipment status, allowing you to move from reactive to predictive.
Part of the system IT-Enterprise product MES provides operational management of production processes, synchronization, coordination, analysis and control of output.