Buildings Alive applies the same machine learning algorithms that we use to identify a building’s potential to expose anomalies in past performance that can point to savings. For example, a building may have experienced similar operating conditions on three days during the past year, but on one of those days the profile of energy use in the morning was lower than the other two. Our technology exposes those differences and quantifies the savings potential. We then work with the building’s management team to investigate and replicate the strategies that deliver the better performance.
The techniques that identify anomalies in a building’s performance history are also used to find the next level of savings by comparing each building’s daily profiles with those of comparable buildings under directly comparable conditions. Even when buildings are on opposite sides of the world and rarely experience similar conditions, often we find that a building in one location is optimised for the most common operating conditions they experience and that significant savings can arise from transferring strategies from buildings optimised for other conditions.
Our services team actively works with customers to help them understand, prioritise and implement changes that our technology identifies. Our software largely automates this process of investigation and root cause analysis in situations where we have direct access to building management system (BMS) data.