Once Properate begins collecting data from a building, it ingests that data into Cognite Data Fusion, which uses AI-powered contextualization services to map relationships between all of the data and create a digital replica of the structure. It uses Bigtable, a NoSQL database designed for handling massive amounts of data, to store and manage real-time sensor readings, energy consumption patterns, and other operational details. Meanwhile, Cloud SQL for PostgreSQL, a relational database, organizes structured information like building schematics, maintenance records, and tenant data. This combination of database technologies allows Cognite Data Fusion® to create a comprehensive and dynamic digital twin that accurately reflects the building's real-world state.

Systems and sensors are mapped in a hierarchical model, showing how they relate to one another, making it easy to compare the building’s assets. They are also mapped in the digital replica of the building, making it simple to visualize their location and understand how they interact with each other and the building. For example, Properate can use the digital twin to model how the air flows through a building, analyze the efficiency of a building’s heating and cooling systems, and identify opportunities for savings.

With Properate, commercial building owners can manage their buildings through one platform, rather than logging into numerous old systems. Properate built its front end using Firebase, which, according to Properate CTO Knut Omang, “is quick and easy for the front-end team to use, and allowed the company to quickly develop a prototype and build an enriched data model.”

When customers log in to the platform, they can use a wide range of dashboards to explore the stream of live data entering the platform and gain a real-time picture of the building. 

“Properate uses this data to give customers the continuous storytelling of their building,” says Huse. “With our algorithms, customers can spot anomalies and trigger alarms to identify problems, and automatically create tasks for the relevant department to fix the problem.” Properate’s models can also help customers save money with predictive maintenance, rather than replacing parts at set intervals. 

When controlling energy usage, customers can model the amount of heating they are using, identify which parts of a system aren’t working properly, and make it right. Building owners can reduce waste, lower their carbon footprint, and save money. These savings are reflected in the data and easily accessed in Properate’s dashboards, simplifying customers’ ESG reporting.

Properate can also work with the energy grid to manage energy usage during periods of peak demand. This helps to lower buildings’ carbon footprints even further by helping the power grid to avoid burning more fossil fuels. 

“With hundreds of buildings, we can act as an aggregator on their behalf, encouraging buildings to reduce their energy usage during periods of high demand,” Huse explains. “With the grid paying us to reduce consumption, we can split those takings with our customers, helping them to earn money for being more energy efficient.”

“We use Google Cloud to run all of these algorithms,” says Huse. “The idea is not to use AI and machine learning for the sake of it, but for a clear purpose. With Vertex AI, we can take cutting-edge technology and use it to achieve something practical and sustainable.”