Field data collection
Connect PLC · SCADA · MES · LIMS with standard protocols and CDC
Manufacturing AX
Connect production, quality, and facility data into a foundation that can be utilized by AI. Transform field decisions and operational improvements into actionable workflows.
OVERVIEW
Connect scattered manufacturing data to accelerate on-site judgment and response.
Manufacturing AX connects data from facilities, production, and quality systems in real time and creates a data base that can track batch unit lineage and the cause of quality issues. It will then be expanded to anomaly detection, quality prediction, and on-site AI Agent.
Events that occur in facilities are linked to quality results, and accumulated data leads to predictions and automation of field work.
Batch Unit Connections
Link process and quality data based on LOT and secure genealogy
Ideal/Quality Prediction
Early detection of equipment abnormalities and quality deviations based on accumulated data
On-site AI utilization
Automation of question and answer, report, approval and response workflow
From equipment to qualityone data flowThis is the basis for manufacturing AX.
CHALLENGES
Data accumulates, but the cause of quality issues is not linked.
If equipment, production, testing, and quality data are separated by system, it is difficult to connect batch unit genealogy and the cause of quality issues.
Data from facilities and business systems are separated.
PLC, SCADA, MES, LIMS and quality systems store data in different formats and cycles, making it difficult to view the entire process flow together.
There is no link between batch-level processes and quality results.
Production conditions, equipment events and test results do not translate to LOT standards, making it difficult to track the extent of impact of quality issues.
It takes a lot of time to find the cause of quality problems.
Because records and documents from multiple systems must be manually compiled, it can take days to determine the cause and related configuration after an issue occurs.
It is difficult to detect abnormalities and quality deviations in advance.
Because equipment and quality data are not connected in real time, it is difficult to detect process abnormalities and OOS risks early.
On-site response and reporting relies on personnel experience.
It is difficult to maintain consistent response speed and quality as cause identification, report writing, approval and action are performed manually.
SOLUTION
Connect field data in real time, Create a traceable quality operating structure.
01
Real-time collection of field data
PLC and SCADA are connected using standard protocols such as OPC-UA, and MES and LIMS are linked based on CDC to minimize the load on the operating system.
02
LOT · Batch unit data connection
Equipment events, production conditions, and test/quality results are linked on a batch-by-batch basis to secure a lineage that extends from process to quality.
03
Batch Trace-based quality tracking
When a quality issue occurs, we trace back related processes, equipment, raw materials and test results to quickly identify the cause and scope of impact.
04
Anomaly detection and quality prediction
Based on accumulated manufacturing data, signs of equipment abnormalities and quality deviations are detected early and response time is accelerated.
05
On-site AI Agent
SAMANDA answers on-site questions and supports cause analysis and response report writing, reducing staff's judgment time.
06
Approval and response workflow automation
Standardize response procedures by linking abnormality notifications, personnel review, approval, action and result records into one workflow.
ARCHITECTURE
From facilities to quality, Connect manufacturing data and AI into one flow.
AkashiQ integrates and refines manufacturing data, Batch Trace and KosmosAI analysis, and SAMANDA field agents complete the quality response flow.
Manufacturing AX Reference Flow
It collects equipment, production, and testing data from the on-premise environment inside the factory, and connects it to batch tracking, quality prediction, and field work automation.
Equipment · PLC / SCADA
OPC-UA based real-time streaming
MES
CDC-based operating system no-load interconnection
LIMS · Quality System
Continuous loading of test results and quality data
MANUFACTURING DATA FOUNDATION
AkashiQ
Equipment, MES, and LIMS data are collected in real time and integrated, stored, and purified. Built-in data lineage and governance by connecting process and quality data on a batch (LOT) basis.
Real-time IngestionCDC IntegrationBatch / LOT TraceData QualityLineageGovernance
Batch Trace
Track batch lineage and quality issue causes from days to minutes
KosmosAI
Quality prediction · Anomaly detection model learning · Serving and AAS asset management
SAMANDA
On-site Q&A, cause report creation and response workflow automation
ORKESTRIX · Factory internal Kubernetes · GPU · AI Runtime · Model Serving on-premise execution base
AkashiQ
Equipment · MES · LIMS data collection, integration, quality, lineage
KosmosAI
Quality prediction/anomaly detection model and AAS asset management
SAMANDA
On-site Q&A, report automation, operational support
ORKESTRIX
On-premises AI execution environment inside the factory
BUSINESS VALUE
In the way of finding a problem after it occurs, Switch to proactive detection and immediate response.
01
Batch Tracking Time Days → Minutes
Quickly identify the cause and impact of quality issues by connecting process, equipment, testing, and quality data on a LOT basis.
02
Detect quality deviations in advance
Based on accumulated data, we secure response time by providing early warning of abnormal signs and OOS risks.
03
Prove audit and regulatory response with data
Systematically manages data lineage, change history, and quality evidence to increase traceability and evidentiary power in regulatory environments such as GMP.
APPLICATIONS
공정과 품질 이력의 연결이 중요한 제조 현장에 적용합니다.
설비 · 생산 · 시험 · 품질 데이터를 연결해 이력을 추적하고 이상과 품질 영향을 빠르게 분석합니다.
제약 · 바이오 제조
설비 · 생산 · 시험 · 품질 데이터를 배치 단위로 연결해 생산 이력과 품질 이슈를 추적합니다.
- LIMS · MES · 설비 · 문서 데이터 통합
- Process, testing, and quality lineage tracking for each batch
- 품질 이탈 위험 감지 · 원인 분석
- Proof of GMP audit response data
REFERENCE · PoC
Domestic pharmaceutical company manufacturing data integration and quality tracking verification
We are conducting a PoC to integrate LIMS and document data, and verify batch-based quality tracking and data usability.
Smart Factory DataManufacturing data integrationMES data analysisbatch trackingPredictive conservation