The BBP’s Building Innovation & Technology Working Group, chaired by Neil Pennell of Landsec, recently brought together industry experts to share their knowledge and facilitate discussions with BBP members on the opportunities presented by Smart technology and Artificial Intelligence (AI). During the meeting we heard a range of perspectives and benefitted from a thought-provoking discussion on how technology might present opportunities for building owners and managers.

Opening the meeting, we heard from Michael Stych, Director, Building Engineering at Arup. Michael presented a fascinating overview of the benefits of Smart buildings and AI, alongside Arup’s approach to transforming buildings to become Smart-enabled. From Michael’s presentation it was clear that Smart technology is not just delivering a step change in energy efficiency and demand-led plant optimisation but is also changing the way that the industry manages and operates buildings more broadly. This includes ‘smart cleaning’ which aligns cleaning frequencies to space demand, reducing food wastage by matching onsite food preparation to daily footfall, as well as ensuring safety by managing and monitoring occupancy levels. He also highlighted the power of the technology to engage occupants and improve user experience, from visualising air quality (in the case of Arup by a cartoon image becoming darker in line with reductions in air quality), to maximising the utilisation of meeting rooms.

It was really interesting to hear from Alan Epps, Operations Director at JLL who shared his experience of developing a Smart strategy and the benefits that can be realised. Alan emphasised the importance of thinking from the outset about the end user and focussing on the outcomes you want to achieve for your building. Alan clearly stressed the importance of approaching the journey in partnership with stakeholders such as IT, onsite engineers & occupants; engaging early and asking the right outcome-focussed questions for the building.

It was inspiring to hear from Michael Grant, COO of Metrikus, about how they had worked with BBP member Bruntwood supporting their digitisation journey. He explained that Bruntwood’s focus has been on automating their buildings as much as possible, as well as using the data and digital learning to anticipate what is going to happen in the building before it happens and empowering tenants to truly engage with their buildings through the visualisation of data.

We then heard two different, but important, perspectives on how to drive performance in buildings:
Firstly, we heard from Andreas Doelle, Environmental Engineer at MeteoViva about how their business is modelling buildings for their clients by creating a ‘digital twin’ (a digital replica of the building which is used to replicate processes, to collect data and predict how the building will perform). Andreas explained how the predictive self-learning, mainly focused on HVAC systems, utilises a vast array of data sets to optimise control of the building creating an autopilot for the daily optimal setting of the BMS. In so doing the temperature and fresh air are otimised according to demand and external factors such as weather. He went on to explain how the system uses feedback and alarms to enable onsite engineers and FM teams to act quickly to address any issues raised when comparing the simulation data with the actual performance of the building. This information can then be used to provide insights to optimise operational building performance. 

Then Andrew Fitzpatrick, Director at BrainBox AI, showcased the concept of autonomous AI and how this can support building performance optimisation and energy efficiency. Andrew described how AI is applied autonomously in a building context from the learning phase where ‘machine learning’ is used to understand how a building behaves based on external factors such as occupancy levels, weather patterns and so forth; to the prediction phase where the system starts to be able to predict what the requirement for the building will be, such as space temperature conditions, based on external factors such as weather patterns (including time of day, angle of sun). 

Andrew explained that by foreseeing undesired events such as dead bands (the crossing of the heating/cooling set points) - the system can autonomously act pre-emptively to adjust the settings accordingly.

Questions and Answers 

Following a short break, we welcomed to a panel Vivek Menon from JLL, alongside those who had already presented. In keeping with the breadth of the topic, the discussion points and questions from BBP members were equally wide-ranging including how to ensure cloud-based cyber security, how to approach interoperability with closed protocol systems and how best to get occupants feedback around how the building is performing.

On the question of lead time to install the equipment the panel agreed that this very much depended on the scope and scale of the project. However, what was clear from all presenters was that when considering a Smart strategy for your sites, whether it be small-scale wireless-controlled sensors to a full-scale AI system, you need to have a strategy and a plan. This strategy needs to consider your outcomes, your stakeholders, commissioning, and maintenance objectives, so it becomes embedded. The panel were quick to emphasise that commissioning is key; and that the setup, adjustment and normalization of the data does take time.

It was clear from the panel that a critical success factor is to get all the key stakeholders around the table from the outset, including IT. The role of the onsite engineers, who may initially be sceptical about how this technology may impact on their roles, was highlighted as pivotal. The building engineers are best placed to truly understand the equipment already on site and how the building works. 

Although there was no doubting the power of the technology from those in the room, the discussion from panelists and members alike focused on the need for there to be an element of oversight and common sense in managing the systems to deliver strategic outcomes. It was observed that people can do things in buildings that are not always logical and that self-learning systems can sometimes learn from poor onsite practices.  The example of a building occupant leaving a window open all night was given. This may ‘trick’ the system into concluding the temperature is getting colder, resulting in a need to put the heating on earlier, when actually, all that is needed is for someone to shut the window. It was agreed that there will always be a need for human oversight, and human reaction to what is happening in the building. 

The topic of naming conventions came up frequently in the dialogue. Clearly there is a need for consistency not just across one’s own estate, but across the industry as a whole – it was agreed we all need to be speaking the same language and create a consistent data dictionary.

Reflections 

The primary focus for convening this roundtable was to focus on the role of Smart buildings and AI to improve energy performance in-use and reduce energy consumption in line with the BBP’s Innovation and Technology Group’s interest in supporting members to reach net zero carbon as rapidly as possible. 

However, what was clear from the presentations was that there is a plethora of other benefits to Smart enabled buildings. It is not just about driving the building to make it energy efficient, it is also about giving the building user the experience they want, it is about the interactions people want to get from their space and the way that the technology itself works in the space. 
It was clear that you need to approach Smart from the perspective of what you want to get from the technology and the use of the data, and then scale your strategy from there. You then need to identify and involve all your stakeholders to develop your Smart building strategy, ensuring the system is specified correctly, configured, and adjusted. This requires engineers with a good understanding and knowledge of the building involved at all stages to ensure you do not get unexpected outcomes that take away the confidence in the systems and what you are trying to achieve.
Buildings are dynamic, and human actions on site will not always be logical. However autonomous the building, it is clear that you will always need some elements of human oversight and intervention to ensure that the building optimisation process gets all the outputs you are seeking to achieve, and priorities such as air quality to support productive work are not negatively impacted. An analogy of washing machines can be drawn here – it is all very well having a highly energy efficient ‘A’ rated machine that washes your clothes using minimum amounts of energy and water, but you may find that at the end of your wash cycle your clothes are still not clean!

Smart buildings and AI are clearly changing the way the industry manages and operates buildings and we look forward to continuing the discussion on this important topic in the future with our members.