Schneider Electric’s Building Graph
To connect data across these various IoT devices and optimize the value of this networked data, Schneider Electric turned to a knowledge graph. Mark Reid, the Director of R&D for Cloud & Mobile Services in the Digital Buildings line of business, knew that the traditional, SQL-based way of looking at the data would not uncover this network effect. They needed to connect the data silos to get a view of the relationships between the data—and they chose Stardog because of our proven track-record with enterprises and the ability to scale limitlessly with our Virtual Graphs.
Mark’s team built a knowledge graph platform called the Building Graph where Schneider Electric gathers buildings data. The Building Graph is a micro-service based architecture where many sources of building data (digitized floor plans, building management system, sensors, elevators, access control, etc.) is linked using Stardog. The system leverages Stardog’s Inference Engine to connect related data and deliver insights to SaaS applications designed to provide valuable customer outcomes.
The Building Graph links many ontologies or data models using Stardog. For building management data, Schneider Electric leveraged the standard semantic descriptions of Brick. With Brick, Schneider Electric could represent their organizational knowledge using industry-standard concepts and relationships between those concepts. This made it easy to ingest data without converting it, because they could store the relationship between Brick and other ontologies in their knowledge graph. For digitized floor plans, the Building Graph leverages GeoJSON to map that building spaces to and from the various other datasets. One key requirement when implementing their knowledge graph was to not require every developer to learn SPARQL, so they created class libraries driven from OWL files to allow developers to work with standard objects instead of learning SPARQL.
Schneider Electric turns to Stardog to integrate IoT data
IoT applications require a shared understanding of the data exchanged between connected devices. Schneider Electric chose Stardog’s Enterprise Knowledge Graph platform for the edge control and collaboration between proprietary systems and third-party systems and devices. The flexible data model also offers quick IoT-device connectivity for faster commissioning and changes.
Schneider Electric uses Stardog to map the relationships between building spaces and a broad set of data points. While the graph contains the relationships, the actual time series information is kept in other storage services. Stardog’s Virtual Graph capability allows the relationships to be tied to the time series data at query time enabling answers to questions like: “Give me temperature readings for the past month in this particular room on this floor,” which additionally can be answered in one query.
Schneider Electric is tracking hundreds of thousands of devices, controllers, and sensors from building management systems, IoT sensors, controls and other systems and software operating a building. A typical day results in at least 25 million events processed in real time.
Increasing business value with better data integration
Schneider Electric leverages enhanced data connectedness from its Building Graph to deliver industry-leading SaaS solutions for customers seeking optimized and efficient building operation at both the site and portfolio level.
The first of these solutions, EcoStruxure Workplace Advisor, helps building facility managers optimize building occupancy and occupant comfort. Workplace Advisor is a suite of flexible digital services to create smart offices through optimized space use, improved comfort and employee experience, enhanced productivity, and reduced facility service costs. Building owners and facility managers derive actionable insights through intuitive dashboards with data from IoT sensors and systems. Plus, employees can connect to their smart offices in efficient, convenient ways using office mobile apps.
Beyond optimization and comfort, safety has become of particular concern, and has taken on new meaning, since COVID-19. The following scenario may become commonplace when workers return to the office: Three employees arrive to their building and are required to select working spaces at a kiosk in the lobby. They want spaces in close proximity that also have good natural light and easy access to a printer. The system finds their desks, and they start their day. A few hours later, the three employees move to a conference room, which is limited to 50% occupancy due to local health regulations, where they will spend the rest of their day. Janitorial staff is immediately notified that the desks have been vacated, and they go to sanitize them.
This whole scenario is only made possible with a knowledge graph. Schneider Electric connects data across sensors, relates this data to particular locations, and, because of Stardog’s inference and ML capabilities, makes recommendations and automates processes.
Another of Schneider Electric’s SaaS solutions, EcoStruxure Building Advisor, is a suite of analytic monitoring services that unlocks a Building Management System’s operational performance so facilities managers can remotely and proactively maintain buildings. It provides key insights into building operations by constantly monitoring systems and identifying faults to proactively address a building’s inefficiencies.
Schneider Electric is well on their way in their knowledge graph journey, leading a digital transformation to create future-ready buildings for premier real estate owners. Mark and his team are building a global platform that links all buildings data, and with 5.5 billion IoT devices in buildings currently and projections that this number will rise rapidly, the sky is the limit for Schneider Electric’s Building Graph.