Artificial Intelligence (AI) and Smart Buildings
The terms Artificial Intelligence (AI) and Smart Buildings are often used in conjunction, but what do they mean? Can a building be smart without AI? What can AI achieve? What insights does it draw? How does it improve efficiency, sustainability and the user experience? We will break these questions down and explore them in detail.
So, first of all, what is AI?
John McCarthy first coined the term artificial intelligence in 1956 to explain concepts around “thinking machines”.
However, modern-day definitions are based around machines that can imitate human behaviour. The English Oxford Living Dictionary gives this definition: “The theory and development of computer systems able to perform tasks normally requiring human intelligence, such as visual perception, speech recognition, decision-making, and translation between languages.”
AI has, of course, been applied to many applications and industries, which has allowed new opportunities to emerge, including improved efficiency, sustainability, wellbeing and comfort, as well as improved insight and security.
We also interviewed Alexandra Moxin, Vice President of Product at Adaptech Group. Alexandra has a background in emerging technology and experience working for the likes of Microsoft; we discussed the role of AI in Smart Buildings. She comments that right now, AI is “hard machine learning; it’s all about taking in data, assessing it and making rapid decisions based on that data.” She goes on to explain the potential within the field of AI and what she finds interesting about the development of it: “I consider what could happen when you have a system that is truly intelligent and can learn and make decisions outside of the existing problem set. Some of the things that are really key as this field develops are the integrity of data, making sure everything is balanced and not biased and also making sure that we’re mindful of the kind of system we’re creating.”
You can listen to our podcast episode with Alexandra here.
There are also many definitions of smart buildings, but Jones Lang LaSalle (JLL) define it as: “Introduce new technology and connect existing systems to enhance the productivity of buildings, workplaces and people. Get access to real-time data to ensure your property or portfolio operates efficiently and creates positive experiences for the people using the space.”
How can AI be applied to buildings?
Building control systems use AI to help make sense of the masses of data collected from the various services and devices within a building.
Alexandra comments on the privacy of smart buildings, highlighting that it is seen as one of the main risks as we incorporate more personal data into buildings, e.g. from our smartphones; we need to be more mindful of this and guarantee privacy for building occupants. We should only collect data that is necessary; many insights can be anonymised; e.g. when gathering data about occupancy, we don’t need to know who is in the room, just how many, where they’ve been and how long they’ve spent in a space.
Another area where AI can be used within buildings is for preventative maintenance. Incorporated sensors that relay data constantly about the ‘health’ of components and devices can allow issues to be mitigated ahead of time. On a basic level, this could flag that a light bulb is approaching the end of life soon; but on an advanced level, preventative maintenance can prevent a building from burning down or flooding, stopping real, harmful damage from being done.
How important is building automation going to be going forward?
For many people, an autonomous building is the ultimate goal, a building that can assist the Facility Manager and Building Occupants acting as an extension of the facilities management team, making decisions that previously required human intelligence. However, this idea of a building that can operate independently will sound quite scary for some other people. There is a balance to be stricken, and this may differ for different organisations; the idea is that AI is being human-like rather than becoming human.
Alexandra believes that automated buildings will become commonplace within smart cities. She also discusses the concept of a “digital twin”, which she describes as “an exact replica of an existing space within a system which uses sensors to gather data so that it can be modelled in real-time”. A digital twin is used by some smart lighting building control systems such as SmartCore for this exact reason, it allows rapid changes to be made and data to be viewed in real-time.
How can AI make the management of a building easier?
IBM answer this question in their report Building Artificial Intelligence into Buildings, “A comprehensive building optimisation system leverages all aspects of building and facility management. These types of systems allow for monitoring the use of space, water, and the usage and allocation of energy. Taking this monitoring one step further, building equipment data collected from IoT sensors that is tagged by location or asset type, and associated with business rules, can trigger algorithms to not only detect but also predict and respond to anomalies. These optimised ecosystems of building technologies identify opportunities for efficiency controls through predictive maintenance. They identify possible root causes so actions can be prioritised, assigned, monetised, and prevented. Recommendations that appear on dashboards or adjustments can be routed directly to the IoT device for action.”
After the building data is collected, the system uses data collected previously as a benchmark to understand what changes have happened; this provides analytics and informs decisions and improvements that should be made.
Artificial Intelligence can provide many benefits for buildings, and we are already seeing it in action across the world. It is constantly evolving and developing, but it is clear the adoption of AI can help us all meet the needs for lower carbon buildings, healthier and safer buildings for occupants, as well as revealing actions and insights many would never have deemed possible a few years ago.