LIVE DEMONSTRATION
UNS Smart Factory
Supervisor Gateway
This dashboard represents a Unified Namespace (UNS) architecture implementation.

Hardware Source: Data is streamed in real-time from an ESP32 embedded gateway simulating high-frequency industrial machine states.

Connectivity: Edge → MQTT Broker → Web Supervisor.

Note: All production data shown is simulated for demonstration purposes and does not represent an operating factory.
Node.js
Socket.io
ECharts
MQTT
ESP32
Invalid Access Credentials
UNS SUPERVISOR | Demo Cell
----/--/-- --:--:--
Day Output (PCS)
Power Load (W)
Cycle Time (S)
NG Rate (%)
📦 Last Piece Trace
WAITING
SERIAL NUMBER--
Model SKU--
Cycle Time--s
Torque--Nm
Angle--°
Diagnostic Code-
📊 OEE Trend (7 Days)Drag Slider
Environment
■ Temp ■ Hum
Dynamics
■ RPM ■ Flow
Avg Yield
98.5%
Avg OEE
86.2%
Energy (kWh)
4,230
Avg CT (s)
32.5
🔴 NG Pareto Analysis
🌓 Shift Performance
💠 Cycle Time Stability
Timestamp Serial No. (SN) Model Result Torque (Nm) Angle (°) CT (s) Error Code

UNS DataLink Engine

Connect Any Device · Deploy Anywhere · Data Freedom

1. Product & Concept Overview

THE VISION

In traditional manufacturing, data is trapped in silos (PLCs, SCADA, MES, ERP). This is the "Integration Mess." UNS DataLink breaks these silos by implementing a Unified Namespace (UNS) architecture.

What is UNS?

It is a single, centralized location where all factory data exists. We structure data using the ISA-95 standard hierarchy:

Enterprise / Site / Area / Line / WorkCell

By acting as a central "Nervous System" (via MQTT Broker), smart devices publish data only once, and any application (Dashboard, AI, ERP) consumes it instantly. No more point-to-point integration nightmares.

Architecture Diagram

2. Core Business Value

ROI DRIVERS
Data Assetization
Pain: Data locked in 'Black Boxes' (proprietary PLCs), requiring expensive fees for API access.
Gain: Return data ownership to you. Clean, standardized data is the fuel for future AI models.
Scalable & Low Cost
Pain: Traditional MES costs millions with slow deployment and rigid structures.
Gain: Decoupled architecture allows "Start Small, Scale Fast." Reuse existing hardware to maximize ROI.
Real-Time Visibility
Pain: Delayed reports. Finding out about scrap rates hours after the shift ends.
Gain: Second-level latency. Shift from "Post-mortem Analysis" to "Active Intervention."
AI Ready Infrastructure
Pain: AI projects fail due to unstructured data and protocol mismatches.
Gain: The UNS structure solves the "Last Mile" connectivity problem, making data Plug-and-Play for ML.

3. Workflow Architecture

HOW IT WORKS
1
Edge Access & Cleaning
Universal Connectivity (Modbus, OPC UA, S7). Edge Governance includes deadband filtering, unit conversion, and anomaly detection to save bandwidth.
2
Unified Transport (MQTT)
Data is published to the Unified Namespace using the Topic structure: Site/Area/Line/Cell. Supports lightweight, bi-directional communication.
3
Contextualization
Raw data is enriched with context (Shift info, Order ID) before being stored or visualized.
4
Full Stack Visualization
Real-time Monitoring (Heartbeat), Analytics (Decision Support), and Traceability (Production Replay) as seen in this demo.