Scholarlink

The implementation of advanced production management systems, in particular digital twins of production processes, allows for the simulation of various process variants in response to changing internal and external factors, which allows for adaptive control of this process. At the same time, building a digital twin of the entire process is most often labor-intensive and troublesome. To reduce this problem, the concept of a distributed modular system was proposed, where modules (digital assistants) are linked to production subprocesses implemented on individual machines or smaller groups of machines. This concept facilitates the construction of a digital twin for complex processes. Large models are built as a multimodel - a non-homogeneous model of a complex object or process, consisting of integrated simpler models of a different nature, which are easier to build and tune to a specific machine/group of machines implementing individual processes. This approach facilitates the integration of IT systems in a manufacturing enterprise, allowing, among others, the use of the synergy effect between robotic process automation (RPA) and artificial intelligence (AI). These technologies are used in the proposed Intelligent Integration and Automation of Information Systems (SIIA ITS), which consists of cooperating digital assistants and a master module that coordinates their work and combines the results of their actions. Digital assistants, which can also act as digital twins of production subprocesses, allow for the simulation of various process variants in response to changing internal and external factors. The solution integrates advanced technologies such as machine learning, artificial intelligence and predictive optimization models, creating a comprehensive tool supporting decision-making processes. The effectiveness of the proposed model of the production process optimization system has been confirmed in practice on the example of a station for machining a cast iron hub, where the test results showed a significant improvement in the efficiency of production processes.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Application of Multimodel Concept and Virtual Assistants to Build a Digital Twin for Supporting Adaptive Control of Manufacturing Process

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Mariusz Piechowski,
  • Ryszard Wyczółkowski

摘要

The implementation of advanced production management systems, in particular digital twins of production processes, allows for the simulation of various process variants in response to changing internal and external factors, which allows for adaptive control of this process. At the same time, building a digital twin of the entire process is most often labor-intensive and troublesome. To reduce this problem, the concept of a distributed modular system was proposed, where modules (digital assistants) are linked to production subprocesses implemented on individual machines or smaller groups of machines. This concept facilitates the construction of a digital twin for complex processes. Large models are built as a multimodel - a non-homogeneous model of a complex object or process, consisting of integrated simpler models of a different nature, which are easier to build and tune to a specific machine/group of machines implementing individual processes. This approach facilitates the integration of IT systems in a manufacturing enterprise, allowing, among others, the use of the synergy effect between robotic process automation (RPA) and artificial intelligence (AI). These technologies are used in the proposed Intelligent Integration and Automation of Information Systems (SIIA ITS), which consists of cooperating digital assistants and a master module that coordinates their work and combines the results of their actions. Digital assistants, which can also act as digital twins of production subprocesses, allow for the simulation of various process variants in response to changing internal and external factors. The solution integrates advanced technologies such as machine learning, artificial intelligence and predictive optimization models, creating a comprehensive tool supporting decision-making processes. The effectiveness of the proposed model of the production process optimization system has been confirmed in practice on the example of a station for machining a cast iron hub, where the test results showed a significant improvement in the efficiency of production processes.