What We Launched With
In January 2022, VX-Olympus launched as an IoT Application Enablement Platform. The core architecture was solid: multi-tenant device management, multi-protocol connectivity (MQTT, HTTP, LoRaWAN, Modbus, OPC-UA, CoAP), Node-RED rule chains for business logic, scalable time-series storage, and a flexible dashboard framework.
The value proposition was clear: stop building IoT infrastructure from scratch. VX-Olympus handled the connectivity, the data storage, the rule engine, and the dashboard framework — the infrastructure that every IoT project needs and few organizations should build themselves.
The early deployments validated this. Cold chain monitoring for a convenience store chain. Water utility AMI across a Kansas municipality. Smart factory monitoring at a precision manufacturing operation. The platform connected sensors, processed data, and generated alerts — reliably, at scale.
We thought we understood what customers needed.
What Production Deployments Taught Us
Eighteen months into production deployments, we started hearing the same things from customers:
“When the alert fires, what do we do with it?”
The alert fires. A maintenance technician gets an SMS: “Press 14B bearing temperature above baseline.” They walk to the machine. They assess the condition. They decide to replace the bearing. They go get parts. They do the work.
Where in VX-Olympus does any of that get documented? Nowhere. The work order was created in the CMMS. Or on paper. Or in someone’s head. The IoT platform had a record of the alert; it had no record of the response.
The next time that bearing showed the same pattern, VX-Olympus had no information about what was done last time. The maintenance tech relied on memory. The platform provided no historical context.
“The sensor shows a problem. But what does that mean for this specific machine?”
A temperature of 92°F means different things for different equipment in different environments. A motor that typically runs at 88°F is showing a more significant anomaly at 92°F than a motor that typically runs at 89°F. VX-Olympus stored the readings; it had no concept of what “normal” looked like for any individual asset.
The alert thresholds were configured by the system integrator at deployment time — absolute values, not baselines derived from the specific asset’s operational history. A year into operation, those thresholds were still the values that had been set on day one.
“We know what the sensor shows. But we don’t know where the equipment is on the floor.”
Sensor data existed in the platform. The asset it was attached to existed in the physical facility. The connection between the two — a floor plan that showed which sensor corresponded to which physical location — didn’t exist in VX-Olympus. Site supervisors looked at device IDs and dashboards; they couldn’t see the state of the facility in spatial terms.
“The data is useful for our site. But we can’t share it with the maintenance contractor.”
Multi-tenant architecture provided data isolation between customer organizations. But sharing data with contractors, sub-contractors, or service partners while maintaining security boundaries required configurations that were more complex than they should have been.
What We Built in Response
The three years following launch were shaped by these four insights. Each became a development priority:
Year 2 Response: Maintenance Integration
The disconnect between IoT alert and maintenance response was the most universally reported gap. The solution: bring the maintenance management workflow into VX-Olympus.
VX-Olympus 2023-2024 added:
- Work order generation from rule chain alerts (automatic, with sensor context attached)
- Work order management: assignment, status tracking, completion documentation
- Asset maintenance history: every work order linked to the asset record
- Parts and inventory tracking integrated with work orders
- PM scheduling based on operating hours (computed from sensor data) rather than calendar time
The change this produced: the loop closed. Alert → work order (automatic) → repair → work order closure → maintenance history updated. The next alert for the same asset arrived with the context of every prior maintenance event.
Year 2-3 Response: Digital Asset Twins
The baseline context problem required rethinking how device data was structured. A device record — an ID with a credential and a data stream — was not an asset model. An asset model required identity, state, history, and derived intelligence.
VX-Olympus’s digital twin layer added:
- Asset identity records (manufacturer, model, serial number, installation date, specifications)
- Baseline computation from 30-day rolling history
- Health scoring from current readings relative to baseline
- Asset relationships (this motor drives this pump, which is part of this production line)
- Alert thresholds that dynamically adjust as baselines shift over time
Year 3 Response: Space and Floor Plan Visualization
The spatial context problem required a visual layer that mapped sensor data to physical locations. VX-Olympus’s space visualization added:
- Floor plan upload and device pinning
- Live telemetry state overlaid on floor plans (color-coded icons showing alert status)
- Multi-floor building views navigable by floor and zone
- RTLS live views for facilities with real-time location tracking
Year 3 Response: Enhanced Access Control for Ecosystem Collaboration
The contractor access problem required a more flexible RBAC model for cross-organization data sharing. VX-Olympus’s enhanced access model added:
- Guest access levels with read-only permissions limited to specific assets or asset groups
- Service partner accounts with access scoped to specific equipment types or facility zones
- Audit logging for all cross-organizational access events
The Platform After Three Years
VX-Olympus in early 2025 is a different platform than the one that launched in 2022. The connectivity layer — the thing it started as — is still there, unchanged and improved. What changed is everything above it:
The raw sensor data now feeds into asset twin records with operational context. Alerts now generate structured maintenance responses rather than just notifications. The floor plan visualization connects data to space in the way operations teams actually think about their facilities. PM scheduling follows equipment condition rather than calendar intervals.
The platform evolution was not a planned roadmap executed on schedule. It was shaped by what customers found missing when they deployed in production — the gaps between “we collect the data” and “we can act effectively on the data.”
What Comes Next
The three-year journey from connectivity platform to operational intelligence platform is the foundation for the next phase. The data models, the maintenance workflows, and the asset twin architecture that exist in VX-Olympus 2025 are the infrastructure that enables what comes next: AI-driven operational intelligence.
The platform that can answer “what is the sensor reading?” is built. The platform that can answer “is this reading unusual for this asset?” is built. The platform accumulating the history to answer “when is this asset likely to need maintenance based on its pattern?” is building that history now.
The AI layer that will deliver actionable predictions from that history is coming. It requires the data foundation to exist first — and the data foundation now exists.
The next article in this evolution story covers what we’ve learned from customers about why device management alone was never enough, and what building the full intelligence layer required.
Talk to our team about what the current VX-Olympus capabilities look like for your specific operation.