.png)
NAU’s CIO Chose Buy Over Build to Turn 1M IoT Points Into Actionable HVAC Control
Northern Arizona University's staff of IT experts was able to successfully access data from thousands of IoT devices; they just struggled to use it effectively.
“We had tremendous trouble storing data in a way that we could operationalize it,” said Steve Burrell, the now-retired former Chief Information Officer at NAU.
The 175-building, 7-million-square-foot campus had multiple BMS platforms, room scheduling systems, meters, and in-room technologies. NAU’s IT team could make API calls and scan networks to get data, but aggregating BACnet and IoT signals in a way that supported real use cases stalled out.
Burrell framed it as a buy-versus-build decision. “I made the choice that we needed to go out and find the experts who knew how to create the ontologies that would support OT and facilities and our sustainability goals,” he said.
NAU deployed a single on-site gateway that observed nearly a million data points. With support from their IDL partner, Mapped, the data is normalized in real time using Brick Schema and then streamlined to the Willow platform via an ontological conversion, enabling seamless data communication between Willow and Mapped.
Improving the relational context within the data enabled occupancy-based HVAC control. By linking room booking data and Crestron occupancy sensors to HVAC zones, NAU sets spaces to standby when both signals show unoccupied. In targeted spaces, HVAC energy fell by around 30%.
As the leader of NAU's IT department, Steve Burrell provides an interesting example of why shopping for data-layer support makes sense in the "buy vs. build" debate. Converged infrastructure and clean data pipelines, which NAU was proficient at, aren’t the finish line. Without an ontology layer that maps systems to physical relationships, scale and control remain out of reach.
Register for the next Nexus Labs event
Sign up for the newsletter to get 5 stories like this per week:
Northern Arizona University's staff of IT experts was able to successfully access data from thousands of IoT devices; they just struggled to use it effectively.
“We had tremendous trouble storing data in a way that we could operationalize it,” said Steve Burrell, the now-retired former Chief Information Officer at NAU.
The 175-building, 7-million-square-foot campus had multiple BMS platforms, room scheduling systems, meters, and in-room technologies. NAU’s IT team could make API calls and scan networks to get data, but aggregating BACnet and IoT signals in a way that supported real use cases stalled out.
Burrell framed it as a buy-versus-build decision. “I made the choice that we needed to go out and find the experts who knew how to create the ontologies that would support OT and facilities and our sustainability goals,” he said.
NAU deployed a single on-site gateway that observed nearly a million data points. With support from their IDL partner, Mapped, the data is normalized in real time using Brick Schema and then streamlined to the Willow platform via an ontological conversion, enabling seamless data communication between Willow and Mapped.
Improving the relational context within the data enabled occupancy-based HVAC control. By linking room booking data and Crestron occupancy sensors to HVAC zones, NAU sets spaces to standby when both signals show unoccupied. In targeted spaces, HVAC energy fell by around 30%.
As the leader of NAU's IT department, Steve Burrell provides an interesting example of why shopping for data-layer support makes sense in the "buy vs. build" debate. Converged infrastructure and clean data pipelines, which NAU was proficient at, aren’t the finish line. Without an ontology layer that maps systems to physical relationships, scale and control remain out of reach.
Register for the next Nexus Labs event
Sign up for the newsletter to get 5 stories like this per week:
This is a great piece!
I agree.