KR-20260065422-A - Method for generating distribution and logistics artificial intelligence learning data using digital twins and its system

KR20260065422AKR 20260065422 AKR20260065422 AKR 20260065422AKR-20260065422-A

Abstract

The present invention is characterized by comprising: a main platform construction step for constructing a main independent platform for AI logistics data analysis using a digital twin; an education and research platform construction step for constructing an AI logistics data analysis and education and research platform using a digital twin after the main platform construction step; and a virtual simulation construction step for constructing a virtual distribution and logistics simulation process using a digital twin after the education and research platform construction step.

Inventors

이용현
신명일

Assignees

(주)원제로소프트

Dates

Publication Date: 20260508
Application Date: 20241101

Claims (10)

Main platform construction phase for building a main independent platform for AI logistics data analysis using digital twin; After completing the main platform construction phase described above, an education and research platform construction phase for establishing an AI logistics data analysis and education and research platform using a digital twin; and A method for generating distribution and logistics artificial intelligence learning data using a digital twin, characterized by including a virtual simulation construction step for constructing a virtual distribution and logistics simulation process using a digital twin after going through the above-mentioned education and research platform construction step.
In claim 1, The above main platform construction phase is, Step 1: Securing a systematic collaboration foundation by enhancing visibility information provided by users; A second step of defining a predictable platform operating system processor after going through the first step above; A third step of linking and collaborating with a partner using a standard integration API after completing the above second step; A method for generating artificial intelligence learning data for distribution and logistics using a digital twin, characterized by including a fourth step of securing operational flexibility, quantitatively monitoring, and securing an analysis basis after going through the third step above.
In claim 1, The above education and research platform construction phase is, Step 1-2: Securing data for learning and research from systems currently operating in actual logistics industry sites; Step 2-2, which involves processing data and analyzing big data through the digital logistics analysis main platform after completing the above Step 1-2, and utilizing it as a research project learning platform; Step 3-2, which involves configuring an information processing platform based on time-series data analysis for transaction data generated in the overall distribution and logistics processes according to the flow of time and processes after completing Step 2-2 above; Step 4-2, which involves receiving and managing source data from trunk logistics operation data to provide continuity of physical events after passing through Step 3-2 above; and A method for generating distribution and logistics artificial intelligence learning data using a digital twin, characterized by including: Step 5-2, which involves building a distribution and logistics data platform and S/W foundation after going through Step 4-2 above, utilizing it as an AI learning platform based on the digital twin and platform, and utilizing it for fostering AI logistics experts.
In claim 1, The above virtual simulation construction step is, Step 1-3: Establishing a system capable of linking and managing transaction operation data from joint participating institutions and external sources; Steps 2-3, which are utilized for practical training and in-house research projects based on operational data after completing the above Steps 1-3; Step 3-3, which involves developing basic technology for a logistics network operation data system after undergoing Steps 2-3 above; and A method for generating artificial intelligence learning data for distribution and logistics using a digital twin, characterized by including: Step 4-3, which involves designing and implementing linkage technology with existing joint research institutions and external information systems after undergoing Step 3-3 above.
In claim 1, The above-mentioned education and research platform construction step is a method for generating distribution and logistics artificial intelligence learning data using a digital twin, characterized by the ability to selectively and automatically generate standard distribution and logistics data that reflects actual environmental information for learning and research.
In claim 1, A method for generating distribution and logistics artificial intelligence training data using a digital twin, characterized in that the above-mentioned virtual simulation construction step generates artificial intelligence training data including data transaction information for data movement objects by automatically extracting processes based on information computed and refined within the digital twin.
Main platform construction unit for building a main independent platform for AI logistics data analysis using digital twin; Education and Research Platform Construction Department for building an AI logistics data analysis and education and research platform using digital twins; A virtual simulation construction unit that establishes a virtual distribution and logistics simulation process using a digital twin; and A system for generating distribution and logistics artificial intelligence learning data using a digital twin, characterized by including: a control unit connected to the main platform construction unit, the education and research platform construction unit, and the virtual simulation construction unit, respectively, and controlling the operation of the main platform construction unit, the education and research platform construction unit, and the virtual simulation construction unit.
In claim 7, The above main platform construction unit is, A collaboration foundation securing unit that secures a systematic collaboration foundation by enhancing visibility information provided by users; A processor definition unit that defines a predictable platform operating system processor; Linkage and Collaboration Department that links and collaborates with partners on standard integration APIs; and A system for generating distribution and logistics artificial intelligence learning data using a digital twin, characterized by including an analysis base securing unit that secures operational flexibility, monitors quantitatively, and secures an analysis base.
In claim 1, The aforementioned Education and Research Platform Construction Department, A data acquisition unit that secures learning and research data from systems operating in actual logistics industry sites; A learning platform utilization unit that processes data and analyzes big data through the main digital logistics analysis platform, and utilizes it as a research project learning platform; An information processing platform component that configures transaction data generated in the overall distribution and logistics processes into an information processing platform based on time-series data analysis according to the flow of time and processes; A data management department that receives and manages source data from trunk logistics operation data to provide continuity of physical events; and A system for generating distribution and logistics artificial intelligence learning data using a digital twin, characterized by including an AI utilization unit that utilizes the digital twin and platform-based AI learning platform after establishing a distribution and logistics data platform and S/W foundation, and utilizes it for fostering AI logistics experts.
In claim 1, The above virtual simulation construction unit is, System construction department for building a system capable of linking and managing transaction and operational data from joint participating organizations and external sources; Project execution unit utilized for conducting practical training and in-house research projects based on operational data; and A system for generating distribution and logistics artificial intelligence learning data using a digital twin, characterized by including a design and implementation unit that develops basic technology for a logistics network operation data system and designs and implements linkage technology with existing joint research institutions and external information systems.

Description

Method for generating distribution and logistics artificial intelligence learning data using digital twins and its system The present invention relates to a method and system for generating artificial intelligence training data for distribution and logistics using a digital twin, which generates research data based on various process designs using information such as actual distribution and logistics environments. Artificial intelligence technology includes machine learning technology, and among machine learning technologies, it includes deep learning technology, which is widely used to analyze images. Deep learning is defined as a set of machine learning algorithms that attempt a high level of abstraction (the task of summarizing key content or functions from a large amount of data or complex materials) through a combination of various non-linear transformation techniques. Deep learning can be broadly viewed as a field of machine learning that teaches computers human ways of thinking. Much research is being conducted to represent data in a form that computers can understand (for example, representing pixel information as column vectors in the case of images) and to apply this to learning (on how to create better representation techniques and how to build models to learn them). As a result of these efforts, various deep learning techniques have been developed. Examples of deep learning techniques include Deep Neural Networks (DNN), Convolutional Deep Neural Networks (CNN), Recursive Neural Networks (RNN), and Deep Belief Networks (DBN). Deep Neural Networks (DNN) are artificial neural networks (ANN) consisting of multiple hidden layers between the input layer and the output layer. While such artificial intelligence technologies are being rapidly applied across various fields, there are still many cases where they are not being properly implemented in industrial settings despite their necessity. With industrial development, the production volume of items in various industrial sectors has increased, and consequently, the number of logistics warehouses capable of storing items scheduled for distribution has also increased. However, losses occurred because the incoming and outgoing quantities of items stored in the logistics warehouse could not be managed, making it impossible to utilize the warehouse efficiently. Therefore, there was an increased need for a method to effectively utilize the logistics warehouse by predicting the outgoing quantities of items stored there. Meanwhile, there is a need for an education and research foundation related to DNA+ convergence logistics technology, the execution and learning of practice-oriented R&D projects based on DNA+ convergence logistics technology, the establishment of a main platform and logistics data analysis necessary for fostering experts based on DNA+ convergence logistics technology, the design of converting various operational and transaction data generated in practice for educational purposes, and the planning and design of logistics data analysis and the main platform. In the past, because it was difficult to prepare actual data, problems arose where the reliability of research results was compromised by forcibly generating arbitrary training and research data, and problems arose where the same data had to be repeatedly loaded when multiple users used it, making it difficult to use for research and training purposes and uneconomical. The following drawings attached to this specification illustrate preferred embodiments of the present invention and serve to further enhance understanding of the technical concept of the present invention together with the detailed description of the invention provided below; therefore, the present invention should not be interpreted as being limited only to the matters described in such drawings. FIG. 1 is a conceptual diagram of a method for generating distribution and logistics artificial intelligence learning data using a digital twin according to an embodiment of the present invention. Figure 2 is an overall flowchart of a method for generating distribution and logistics artificial intelligence training data using a digital twin. Figure 3 is a sub-flowchart of the main platform construction phase. Figure 4 is a sub-flowchart of the education and research platform construction phase. Figure 5 is a sub-flowchart of the virtual simulation construction phase. FIG. 6 is a drawing illustrating a first embodiment of a digital platform for learning and researching convergence technology. FIG. 7 is a drawing illustrating a second embodiment of a digital platform for learning and researching convergence technology. Figure 8 is a diagram illustrating a distribution and logistics process. Figure 9 is a diagram illustrating a data utilization service processor. FIG. 10 is a block diagram of the overall components of a system for generating distribution and logistics artificial intelligence learning data using a digital twin according