Ever since we conducted our first experiments and built our first prototypes almost a decade ago, Buildings Alive has been saving people time. Sure, our main purpose is to help building owners, operators and occupiers achieve their energy and environmental efficiency goals. But for many, their most precious resource is time.

Do you have more control over your time than you did a decade ago? Not likely. And if you manage a building (or portfolio of buildings), you’re probably wondering: “do I even have time to read this?” Please, bear with me!

We all know technology can be the solution; but too often it’s the source of the problem. The proliferation of advanced building systems and sensing and control technologies is driving massive growth in data volumes which, in turn, brings complexity and sucks time. When more technology is overlayed on top of these systems, turning raw data into ‘information’, ‘information overload’ is the inevitable consequence.

The difference between ‘faults’ and ‘anomalies’

Fault detection and diagnostics (FDD) is the umbrella term for algorithms that use building operational data to identify the presence of faults and (in some cases) isolate their root causes. Faults can account for a large proportion of a building’s energy usage so it’s important that building operators and technicians know about them so they can be addressed. But not all faults are equal! Some faults matter a lot, and some are a waste of time. Furthermore, not all faults are necessarily ‘faults’ at all. Because FDD relies on ‘rules’, and those rules are developed and implemented by humans, they are not necessarily ‘smart’ or correct.

To be ‘smart’, a technology needs to learn, adapt and decide.

That is why Buildings Alive has been pioneering a new category of building optimisation analytics called Automated Anomaly Detection and Diagnostics (AADD). Our first step towards AADD was launched in 2017 when we enhanced our Rapid Efficiency Feedback (REF) for building operators to quantify their building’s ‘best ever’ performance under comparable operating conditions and also expose its improvement ‘potential’. REF pinpoints issues and opportunities within 15-minute windows with evidence-based judgements derived from advanced machine learning technology applied to utility data at (usually) the whole building and base building level.

While REF is invaluable for understanding how well a building is doing and how much better it can do, the question of ‘what to do?’ has until now been left up to the interaction and dialogue between building operators and technicians and our expert engineering team. That’s why we are now excited to launch REF+. REF+ is an upgrade to REF because it highlights the key actions that should be implemented and maintained to consistently achieve a building’s optimum energy and environmental performance.

If something matters, REF+ exposes it. If it doesn’t, it won’t!

REF+ uses the judgements created by the data models that underpin REF to learn what operational settings deliver improvements, and what anomalies may be causing problems. Information about operational settings is derived from data extracted from the Building Management System (BMS), including fine-grained sensor and control point data.

Why does this matter?

REF+ learns the causes and consequences of all operational changes that impact building performance and highlights the things that are most notable – good and bad. And because it doesn’t rely on pre-set ‘rules’ determined by humans, its judgements and recommendations are always statistically valid. REF+ answers the question: ‘what are the most effective actions I can take to improve building performance?’ Its focus is timesaving, cost-saving and it is very ambitious! Furthermore, it pinpoints areas for focus so the recipient of the message (typically an FM) does not need deep technical expertise or to spend lots of time trying to understand an issue in detail.

REF gives clear, timely feedback about building performance; REF+ helps answer the logical follow-on questions about what to do? why do it? and how to do it?

If you would like more information about REF+ or a demo, please get in touch!