로고 로고

SITE MAP

REFERENCE AI·Big Data

REFERENCE

Based on over 20 years of experience in building smart factories,
DL Information Technology Co. Ltd. is creating manufacturing big data analysis
and AI application cases in various industrial fields

AI/Big Data Construction Case_ Customized AI Simulation Model Application through Materials, Parts and Equipment-Linked Digital Twin/XR Verification Research and Development Project

In 2024, DL Information Technology Co., Ltd. provided customized AI-based simulation models for Chungbuk root companies through Materials, Parts and Equipment-Linked Digital Twin/XR Verification Research and Development Project

Raw data Analysis environment/Technology Data collection/Loading Visualization

Raw data

  • H/Shaft Production History Data

  • H/Shaft Heat Treatment Inspection Results Data

Production plan and production history data of products manufactured in the H/Shaft line and outer ring line
Heat treatment inspection results and defect determination data of the heat treatment process

Analysis environment / technology

  • DiffusionAE Analysis Model Structure

  • DiffusionAE Performance

DiffusionAE for multivariate time series outlier detection appropriate for One-Class SVM model vs. time correlation analysis Model application
Manufacturing process data preprocessing → Data matching → Missing and outlier detection and removal → Data normalization Building a product quality prediction data set for data analysis through (normalization)

Data Collection / Loading

  • Structure diagram of Pub/Sub-based data collection and linkage

Producer configuration appropriate for data format to accommodate various types
of data
Collection of real-time/periodic/batch data for legacy systems and production equipment in consideration of manufacturing process characteristics

Visualization

  • Structure diagram of Pub/Sub-based data collection and linkage

Support decision-making of field workers and managers by visualizing cause analysis of product defects occurring in the manufacturing process and prediction results
Intuitive visualization of real-time manufacturing process data and defect prediction results through linkage with digital twin system
Paint Point

  • Absence of management system

  • Occurrence of
    defective products

  • Unable to track and manage

  • Delay in response
    to issues

Economic and systematic losses due to manual data management
Absence of manual data analysis process and management system
Occurrence of defective products due to quality control relying on the operator's eyes and experience
Data loss and inability to track and manage due to insufficient data management and management system
Delay in decision-making and response to issues caused by insufficient linkage system between systems
Introduction effects
  • Visibility and immediacy of process information
    Improved standardization and management system for IT utilization work through linkage of legacy system data
    Real-time visibility and safety of production information by monitoring real-time production and facility operation status
    Quick decision-making through rapid information processing, improved user convenience, and sharing of field data
  • Establishment of an efficient production management system that enables real-time confirmation
    Support for rapid decision-making by grasping the entire process progress in real time
    Support for remote work by checking the on-site situation in real time from the office
    Able to respond to problems/abnormalities such as real-time equipment RUN/IDLE/STOP
Client company