We recently completed a project integrating real-time machine status from our CNC machining center into Honeywell MES shop floor monitor. The goal was reducing downtime by catching machine issues faster through automated alerts.
Our implementation connected an OPC UA server running on the machine controller to the shop floor monitor dashboard. We configured status feeds for critical parameters including machine state (running/idle/error), spindle load, and coolant temperature. The alert rule engine now triggers notifications when machines enter error states or when performance metrics fall outside normal ranges.
The integration reduced our response time to machine failures from an average of 45 minutes down to under 5 minutes. Operators receive instant alerts on their shop floor tablets, and supervisors can view real-time status across all 12 machines from a central dashboard. Happy to share our configuration approach and lessons learned.
The alert configuration was done entirely through the standard shop floor monitor interface without custom code. We created rule templates for each machine type with threshold-based triggers. For example, if spindle load exceeds 95% for more than 30 seconds, or if machine status changes to error state. The trick was tuning the thresholds during a two-week pilot to avoid alert fatigue. We started conservative and gradually tightened based on actual operating patterns.
Great questions. The biggest challenge was actually organizational, not technical. Getting operators to trust and act on automated alerts took change management effort. Technically, the most time-consuming part was mapping all the OPC UA nodes correctly and testing edge cases.
For network architecture, we implemented a segmented approach that satisfied both operational needs and IT security. Here’s our configuration:
Network Segmentation:
- All machines connected to isolated production VLAN (10.50.x.x)
- OPC UA servers on machines communicate only with MES gateway server
- MES gateway acts as data diode - one-way flow from production to enterprise network
- Shop floor monitor runs in DMZ zone with controlled firewall rules
- Operator tablets connect via secure WiFi with limited scope
OPC UA Security Configuration:
We enabled OPC UA security mode SignAndEncrypt with certificate-based authentication. Each machine has its own certificate trusted by the MES gateway. This prevented unauthorized access while maintaining real-time performance.
Alert Rule Implementation:
For alert configuration, we built a three-tier structure:
- Critical alerts: Machine error states, safety issues - immediate notification to operators and supervisors
- Warning alerts: Performance degradation, approaching maintenance thresholds - notify operators only
- Informational alerts: Status changes, shift transitions - logged but no active notification
The shop floor monitor integration uses standard OPC UA client connections configured in the resource management module. We created a custom dashboard template that displays machine status with color coding (green/yellow/red) and real-time KPI widgets.
Key Configuration Steps:
- Define OPC UA namespace and node structure for each machine type
- Configure MES gateway OPC UA client subscriptions with 1-second polling
- Map OPC UA nodes to shop floor monitor data points using configuration XML
- Build alert rule templates with threshold logic and escalation paths
- Create operator dashboards with machine status cards and alert queues
Lessons Learned:
- Start with 2-3 pilot machines before rolling out to entire floor
- Document the exact OPC UA node paths - they become critical for troubleshooting
- Set alert thresholds based on actual production data, not theoretical limits
- Include maintenance team in alert routing - they often spot patterns operators miss
- Budget time for operator training - the technology works, but adoption requires buy-in
Our ROI calculation showed payback in 7 months primarily from reduced unplanned downtime. The real-time visibility also improved our OEE by 8% because supervisors could rebalance work when machines went down.
Happy to share specific configuration files or answer more detailed questions about any part of the implementation.
This sounds like exactly what we need! We’re running HM 2023.2 as well with about 8 injection molding machines. How did you handle the OPC UA server configuration on the machine side? Did you use the built-in OPC UA server from your CNC controller or did you need middleware?
We were fortunate that our Fanuc controllers had native OPC UA support. We enabled the OPC UA server feature and configured the node structure to expose the parameters we needed. For older machines without built-in OPC UA, we used KEPServerEX as middleware to convert Modbus TCP data to OPC UA format. The key was standardizing the node naming convention across all machines so the shop floor monitor could map them consistently.