From the BuildingIQ perspective, an AI-driven inference engine is a powerful tool related to the understanding the relationships amongst data and devices in a system using imperfect data. What do I mean by “imperfect data?” Simply that when we don’t have a lot of data or all of the necessary pieces to come to a conclusion about a system, an inference engine helps us to fill in any missing gaps. It helps us identify relationships that may not be obvious, or in the design of the system, and figure out what’s going on inside of that data system without actually having all of the needed information or documentation. It’s not that the designed-in rules are unimportant, it’s that the relationships uncover unintended rules that could be very important!
The applications for such engines span the gamut from consumer to industrial applications. Interestingly enough, in the buildings industry, we don’t really see much if any use of inference engines yet. We see machine learning used extensively in modeling and predictive analytics wherein machine learning is commonly used to look for data pattern changes. The algorithms behind the machine learning are trained to identify what’s normal versus abnormal based on such patterns.
But, as far as inference engines go, BuildingIQ is at the forefront of this movement. We are now able to look at relationships amongst building data, at the device level or at the data point level, comparing both to each other. For example, when looking at a whole building, we can look at its energy consumption as a key data point and thus weight all of the data points and how they move in relation to that whole energy metric. We’re looking at the relationship amongst data, in this instance relative to a key metric, as opposed to a specific change in data.
Epiphany is our inference engine. We’ve chosen to build it as a set of interrelated tools can be used collectively or individually. In every way, it’s a bit of foundational technology for us.
The first example is with our Outcome-based Fault Detection (OFD) solution, which utilizes the Epiphany Engine for some key analytics and root cause identification. We seldom have perfect access to data and Epiphany lets us look at data contextually unaware of rules and how things are put together. Because of that, the OFD service enables our network operators to identify systematic issues that are impacting operations parameters set in place, like occupant comfort and energy consumption. Once issues are identified, they are ticketed and fixed.
Another great value we’re already seeing when applied to building operations and specifically for facilities management is the ability for Epiphany to become a resource to gain a deeper understanding of a building. Knowledge transfer is traditionally passed from generation to generation. But there is a wealth of institutional knowledge that is very difficult to transfer. As a building’s database grows with our services, we’re facilitating the spreading of knowledge about a building that will help facilities teams to be more effective. This kind of knowledge is really powerful for the day-to-day operations of a building. Having an understanding of the way equipment functions, interacts and is interdependent can be really useful for higher-level planning purposes. And having a way to store and transfer that knowledge over time is a major de-risking for building owners.
Steve Nguyen is VP of Product and Marketing at BuildingIQ. He loves products and ideas that transform markets or society. Whether they are transformative in and of themselves, or because they are enablers. He’s driven by creating the stories, teams, and strategy that make these agents successful.