Skip to content

Smart Factory Control Center

Real-time OEE war room for discrete manufacturing

LIVE

This dashboard demonstrates IoT DC3 in discrete manufacturing: a real-time OEE war room combining a machine status matrix, a six-big-loss waterfall and process flow topology to track throughput, yield and equipment pace. All data is collected and normalized by IoT DC3's multi-protocol drivers, with edge-to-cloud delivery and alarm linkage.

Dashboard highlights ​

  • Equipment status matrix: running, idle, fault and down states for the whole line on one screen.
  • OEE ring and six-big-loss waterfall: availability, performance and quality decomposed step by step.
  • Process flow topology: live routing from raw material to finished goods with key cycle times.
  • Output and work-order completion rankings across shifts and lines.
  • Tank levels, trends and an alarm/event log linked for root-cause tracing.

Business value ​

  • OEE moves from after-the-fact reporting to a real-time war room with losses pointing at improvement actions.
  • Alarm-to-status linkage cuts unplanned downtime.
  • Output and yield data is retained automatically for lean improvement and quality traceability.

How to build this on IoT DC3 ​

Line PLCs, sensors and SCADA connect through Modbus, OPC-UA, MQTT and other pluggable drivers, then map onto the unified device-point model (value + quality + timestamp). The rule engine raises alarms on thresholds and durations, while time-series storage keeps history for trend analysis. See the documentation and the GitHub repository.

FAQ ​

Is the data real? ​

This page runs on mock data to demonstrate the visualization and metric definitions; once real devices are connected, the data pipeline is identical.

What if my protocol is not in the driver list? ​

Drivers are pluggable: beyond the 36 built-ins, IoT DC3 supports custom TCP/UDP driver development — see the driver-development guide in the docs.

IoT DC3 · Connect the Physical World to AI