Skip to main content
PILETA Co., Ltd.

Enterprise Data Integration · Digital Thread

From materials testing to life prediction, one digital thread

An enterprise-wide data platform that connects materials testing, microstructure AI, failure analysis and master data through single sign-on (SSO) and a unified repository (PostgreSQL, MinIO).

Data Pipeline — 4 Steps

Digital thread data pipeline

One continuous flow instead of disconnected systems. Data moves seamlessly from collection to prediction.

01

Test data collection

Collects raw materials test data into a unified repository.

02

Materials analysis

Quantitatively analyzes material condition with microstructure AI.

03

Damage assessment

Assesses flaws and degradation through the failure analysis workflow.

04

Life prediction

Predicts remaining life and creep life from accumulated data.

Platform Components

Four products that make up the digital thread

PiletaDAS

Damage Analysis System

A platform for failure analysis and hierarchical management of projects, specimens and tests.

  • Three-layer Project → Specimen → Test data model
  • Failure analysis workflow / automated report generation
  • Dual storage structure that separates the database server (raw data) from the file server (file data), plus an integrated data authentication framework

PiletaMDM

Master Data Management

Integrated master data management / a single source for codes, standards and certifications.

  • Standardized codes for materials, tests and standards (KS, ASTM, EN)
  • Single source of master data for all enterprise applications
  • Version and history management, plus a change approval workflow

PiletaSSO

Single Sign-On

JWT-based unified authentication.

  • Centralized security through single-account, permission-based access control

Creep DL

Deep Learning for Creep Life Grade Classification

Automatic creep damage level classification based on microstructure images.

  • Microscope image input (scanning electron, optical or laser microscope) → preprocessing (brightness and contrast adjustment, noise removal) → general-purpose segmentation with MatSAM → quantification of nine microstructural features → three machine learning models (random forest, XGBoost, support vector machine) → damage level 1–6 with a confidence score
  • Research results: 98.3% classification accuracy with a convolutional neural network (CNN), Vision Transformer (ViT) transfer learning (robust with small datasets), and prediction confidence intervals from a Bayesian neural network (BNN)
  • Feeds directly into creep life prediction and remaining life assessment

Integration Impact

Integration KPIs

Key performance indicators for enterprise-wide data integration.

70%+Processing speedReal-time integrated processing
100%Standardization rateUnified master data
12+Integrated systemsEnd-to-end digital thread
Decision-making speedRapid, data-driven response

Key Differentiators

Three platform differentiators

01

Unified authentication with PiletaSSO

One account and permission-based access control across platform applications, with security managed centrally.

02

Master data standardization

Ensures data consistency by unifying coding systems across the company.

03

Extensible architecture

Flexible integration of new systems and data sources.

Platform Architecture

Four-layer architecture

Authentication and master data, applications, storage and the data pipeline are integrated into a single stack.

AUTHENTICATION & MASTER DATA

  • PiletaSSO
  • PiletaMDM

APPLICATION SYSTEMS

  • PIMP (PILETA Intelligence Materials database Program)
  • Creep DL
  • PiletaDAS
  • Nondestructive testing (NDT) management
  • Life assessment

DATA STORAGE

  • AI-optimized database (PostgreSQL)
  • Large-capacity file server (MinIO)
  • Unified repository

DATA PIPELINE

  • STEP 01 Test data collection
  • STEP 02 Materials analysis
  • STEP 03 Damage assessment
  • STEP 04 Life prediction

Benefits & Deployment

Benefits and deployment

Benefits

  • Eliminates data gaps between testing, analysis, assessment and prediction
  • Stronger audit readiness through unified standard codes (KS, ASTM, EN)
  • Turns training data for AI models (Creep DL) into a reusable asset

Deployment

Developed in-house and operated as the company-wide data integration system of PILETA Co., Ltd., the platform is being applied to new projects.

Ready When Reliability Matters

When integrity needs proof, PILETA delivers

The PILETA Test & Evaluation Center is accredited by the Korea Laboratory Accreditation Scheme (KOLAS) as a testing laboratory (Accreditation No. KT921). From testing and failure analysis to digital solution consultations, our engineers respond to you directly.

T. +82-42-368-0180 · pileta@pileta.co.kr