◎ Jiao Qingguo and Sun Hongxi, Qingdao Wongoing Information Technology Co., Ltd.

I. Concept of Digital Twin

“Digital twin” refers to digitally copying physical objects, simulating the behavior of objects in real environment, and performing virtual simulations on products, manufacturing processes, and even entire factories, thereby improving the production efficiency of product research, development and manufacturing of manufacturing enterprises.

“Digital twin” helps enterprises optimize, simulate and test in a virtual environment before they actually go into production, and they can simultaneously optimize the entire enterprise process during the production process, ultimately achieving efficient flexible production, rapid innovation and market launch, and forging lasting competitiveness of enterprises.

“Digital twin” can support enterprises to integrate and digitally transform their entire value chain, to create a consistent and seamless data platform for all links from product design, production planning, production engineering, production implementation, to service, and to form a model-based virtual enterprise and a mirror image of real-world enterprise based on automation technology. The digital twin model has the characteristics of modularity, autonomy, and connectivity. It can break the barrier between reality and virtuality from the perspectives of testing, development, technology, operation, maintenance, etc., and achieve highly digitization and modularization of production, management, and connection throughout the full product life cycle.

The new management mode of digital twin, can realize the unity of physical manufacturing and virtual manufacturing, create a world of information space, and realize the digitalization, networking and intelligence of manufacturing, reengineering business manufacturing processes in the information space and continuously improving manufacturing efficiency based on this. The new management mode will be the manufacturing based on cyber-physical system, the manufacturing which is data-driven, software-defined, platform-supported, and the manufacturing where physical manufacturing and virtual manufacturing interact in real time. Whether product, equipment, or process flow, it will appear in digital twin. Through the application of virtual manufacturing such as digital twin, it can realize the process transformation from fragmentation to integration, from local to global, and from static to dynamic, covering the entire process of R&D design, manufacturing process, and service operation.

Digital twin is also one of the necessary technologies for the development of CPS.

II. CPS: Necessary Technology for Information System Development

CPS is the core technology of Industry 4.0. CPS-based smart equipment and smart factories will lead the transformation of manufacturing methods.
CPS builds a closed-loop channel for the data interaction between information space and physical space, which can realize the interaction linkage between the virtual information and physical object. Based on the static model generated by the physical object modeling, through real-time data collection, data integration and monitoring, the working status and work progress of the physical object are dynamically tracked (such as collecting measurement results, tracing information, etc.) All elements of the physical object in the physical space are reconstructed in the information space to form a “digital twin” with abilities to perceive, analyze, decide, and execute, which is also known as digital mapping, digital mirroring, and digital twin. At the same time, with the help of the ability of information space to comprehensively analyze and process data, an effective decision on external complex environmental changes is formed, and affects the physical object in a manner that “the virtual controls the real”. In this process, physical object interacts with virtual information, working together to improve the efficiency of optimal resource allocation.

CPS supports the deep integration of informatization and industrialization. Through the integration of advanced information technologies, including sensing, computing, communication, control, etc., and automatic control technologies, a complex system is built where elements in physical space and information space timely interact and efficiently coordinate with each other, such as human, machine, object, environment, information, etc., realizing on-demand response, rapid iteration, and dynamic optimization of resource allocation and operation within the system. The characteristics of CPS are shown in Figure 1.

Fig. 1: Characteristics of CPS

The intelligent implementation of CPS is roughly divided into four stages. The first stage is the autonomous perception towards system environment information; the second stage is that after the sensing information is obtained through sensor network, CPS performs appropriate processing on the acquired information, such as rejection of useless information, classification of information, etc.; the third stage is to model the overall system of CPS based on the establishment of database to complete the cognitive task; and the fourth stage is to realize final decision and system control through the overall model and database . As shown in Figure 1, four stages in the center are the step-by-step design of CPS. Three technologies in the ring are the technologies that need to be considered in the overall system design, and the digital twin is the most critical technology.

III. Features of Digital Twin Management Mode

With the application of digital twin management mode, enterprises can achieve the following transformations:

1. Virtualization. The digital mirror image built by digital technology which is symmetrical to the manufacturing can simulate and show the production site, and show the actual status of field production in real time. Some operators and management personnel can understand and operate the on-site production equipment without going to the site, therefore realizing the conformity of the virtual and the real.

2. Virtual manufacturing. Product R&D, manufacturing, operation and maintenance during the manufacturing process are transformed into relatively low-cost digital informa-tion in the virtual world, and collaboration and model optimization are performed to improve the efficiency of research, development and manufacturing. Through the virtual-real connection between the “twins”, the data is continuously iterated, the model is continuously optimized, and then the optimal solution is obtained.

3. Virtual debugging. Virtual debugging can be performed at the beginning of product design. The completion of debugging means that the entire production line and the entire system are considered to meet the standard specifications. Then the manufacturing starts.

4. Sample design. In traditional manufacturing mode, the engineer needs to make a sample after finishing the design. The sample may have many defects, and then design changes will be made targeting at these defects. Batch production can only be carried out after improvement. This process is time-consuming and labor-intensive. However, with the digital twin and its virtual simulation system, this step can be done based on software. After the system has certified and tested every detail, it is put into production at one time, shortening the R&D and testing cycle of new products.
Digital twin is the foundation of intelligent manufacturing system, with characteristics of modularization, scalability, fidelity, and connectivity. It can break the barrier between reality and virtuality from the perspectives of testing, development, technology, operation, maintenance, etc., and achieve highly digitization and modularization of production, management, and connection throughout the full product life cycle. “Digital twin for product”, “digital twin for production technical process” and “digital twin for performance” can completely and truly reproduce the entire enterprise.

Enterprises can optimize, simulate and test in a virtual environment before they actually go into production, and they can simultaneously optimize the entire enterprise process during the production process, ultimately achieving efficient flexible production and rapid innovation and market launch.

IV. Management Model Based on Digital Model

Models are the key to building digital twin. Enterprises should build the optimal model according to their own manufacturing.

The production and manufacture of rubber tires can be realized based on digital models. Through “assignment” and “enablement”, digital twin for the production of rubber products can be established, laying a solid foundation for the intelligent construction of enterprises.

Model-based definition requires that all types of information defined by the product shall be organized in a model-based manner. By building an appropriate enterprise model, a mapping is created in the digital model of virtual world and the physical object of real world. Through model-based R&D and manufacturing, digital twin for enterprise products, digital twin for production processes, and digital twin for equipment performance, etc. are established to serve the management of products in the entire life cycle of research, development, production, sales, and marketing. Digital twin for product R&D can achieve simulation and verification of research and development; and digital twin for manufacturing can achieve simulation of production, logistics and manufacturing, and process quality testing, to produce the best products with the best  efficiency.

The R&D applications in tire enterprises include model-based product design, pattern design, formula design, structural design, and performance analysis models.

The production and manufacture in tire enterprises includes model-based construction standards, work instructions, planned production schedules, manufacturing execution, and quality inspection.

V. Analysis of Applying Benefits

The applying of digital twin management model can bring benefits to the production, manufacturing and R&D of enterprises, promote the transformation and upgrading of enterprises, and improve the profitability of enterprises. The analysis on tire manufacturing benefits is as follows:

1. Optimize Production Planning

Virtual planning is carried out for the entire production line, from mixing, rolling, cutting, forming, vulcanizing, to final quality inspection, and logistics planning is made upon equipment, logistics, and production. By virtualizing production, logistics, etc., this can optimize the manufacturing process, reduce time waiting for materials, and realize automation and intelligence of production logistics.

Fig. 2: Schematic Layout of Automatic Logistics in Smart Factory

2. Visual Control

By digital twin for production, the entire production process can be planned in a fully virtual environment, from layout design to visualization of material flow and production. Virtualized production helps test and optimize new production lines, virtually debug production, and virtually optimize logistics. At the same time, it can also provide virtual guidance for the construction of new factories, monitor the current production status in real time, and raise alarms for abnormalities.

Fig. 3: Automated Logistics Visualization in Factory

3. Digital Operation and Maintenance

Through digital twin mode (see Figure 4 for digital operation and maintenance), it can remotely monitor, operate and maintain the running conditions of equipment production line and the key components of the equipment. Fault detection, health management and predictive maintenance of equipment can be achieved without the need to be on-site, and remote control and operation & maintenance management can be achieved if necessary.

Fig. 4: Digital Operation and Maintenance

4. Digital Twin Management The digital twin management

solution can greatly reduce the R&D and testing cycle of new products, and can predict the performance of production units and products, shortening the time to come into market. It can show the development status of product or factory in the entire life cycle, enabling operators to predict behavior, optimize performance, and draw on experience of previous design and production.

Great value can be gained by using digital twin to perform “forecast” scenarios and predict future performance. The ultimate goal of digital twin is to achieve a closed-loop connection between the virtual environment of product development and production planning, and the actual production system and product performance. With this connection, it can achieve actionable insights into the actual production environment to make informed decisions throughout the life cycle of products and production operations.