open search menutoggle menu

Smart building and data to meet climate challenges

At a time of heatwaves and increased volatility of climate events, in a context where energy costs are soaring, making buildings intelligent and capitalising on their data is becoming imperative. First, it is a way of making them more sober. But it is also an elegant way to reduce our carbon footprint massively and effortlessly...

By Eric Gacia and Antoine Nguyen from the Data & AI Advisory practice in France

Real estate: an energy-intensive sector

Collectively, buildings in the EU are approximately responsible for 40% of our energy consumption and 36% of greenhouse gas emissions, which mainly stem from construction, usage, renovation and demolition.[1]Building eco-conscious buildings seems an important lever in the fight against climate change.

Energy savings naturally involve improving insulation on new builds or renovations. But energy efficiency also comes from use. As such, the smart building is a significant lever that allows it. 

In real estate, it is now possible to use numerous data sources to monitor, control and modify their behaviour according to their activity. Thus, it is now possible to control ventilation, air conditioning, heating, shutters, and other systems, depending on the input data. 

A building is considered smart when connected devices can communicate with each other.

IOT sensors and data: a prerequisite for making buildings smart.

One of the challenges of the smart building is to collect data and to report it regularly. Each measured parameter requires its connected object (IoT - Internet Of Things), capable of transmitting information over the network in real time or over a longer time. Other connected objects must perform the reverse process (actuators) to respond to changes in measured parameters. For example: sensors measuring brightness and outside temperature make it possible to operate shutters that automatically close on the faces of a building exposed to the sun in the middle of summer (real time), sensors that measure air pollution to adapt the ventilation of the premises (long time).

It is thus possible to measure different parameters within a building and act accordingly. 

Controls can be carried out on an ad hoc, planned or automated basis in order to optimise the building’s life parameters. 

The IoT opens the way to buildings that respond, in real time, to the use and needs of people to adapt supply as closely as possible to demand.

All these digital inputs are a revolution for Technical Building Management (BMS), thanks to the data collected but also thanks to artificial intelligence, which makes it possible to refine prediction techniques by enabling the analysis of a large quantity of data at the same time. 

With the emergence of the IoT, it is possible to view at any time all the parameters of a building in real time but also its state soon.

This is called a digital twin. Measuring and visualising consumption flows is one of the first pillars towards optimising energy consumption. To do this, we need to develop tools for monitoring consumption on a smart building. 

In the near future, there will be more flows: electricity consumption, recharging of electric cars that consume energy on one side and then solar panels, wind turbines and electric cars that inject electricity on the other. We will need to be able to monitor all these flows to better manage them. 


The challenge: intelligently cross data to maximise frugality

There is a distinction to be made when talking about smart buildings. Depending on the type of use, there will be no need for the same "intelligence". A residential building does not live in the same way as a tertiary building, whether it is offices, a gymnasium, or a factory.

It is easier to simulate the use of a tertiary building because the rules of use are already defined and more settled. Offices will probably be empty between 20 and 8 am. This makes it easier to schedule technical management without using AI.   

AI optimization in residential buildings is still possible, however, with habit recognition that will be more accurate and adaptive than programming in advance. It is possible to consider controlling the ventilation according to the occupancy rate of the offices, automatic extinguishing of the lights thanks to motion sensors, intelligent waste management with filling detectors. 

Where artificial intelligence will have the greatest impact will be on optimising equipment maintenance. With the installation of sensors, we obtain information about their wear and performance and thus it will be possible to ensure predictive maintenance without the need for regular checks by an agent. It therefore allows a rapid reaction and limits yield losses as well as repair costs. 

In all cases, savings in the tertiary sector will be easier to achieve due to the scale of the number of people impacted compared to the number of people in charge of the BMS.

The residential sector is currently not very automated. For now, this boils down to the scheduled start-up of water heaters and the possibility of delayed departure of household appliances. These automations are possible, but not systematically implemented by individuals: due to a lack of knowledge, interest or resources. 

It should be noted that during containment due to the Covid pandemic in 2020, residential consumption did not change significantly whereas in normal times, the presence at home is 15%. During this period, for instance in France, household final energy consumption increased by only 3% compared to 2019 (corrected temperature deviation). This means that consumption is not suitable for real use. In France, the Minister for the Energy and Solidarity Transition also stressed that this “shows that heating devices operate in normal times, including when residents are not at home. The stability of energy consumption shows that there will be significant levers for reduction". Intelligent and autonomous management through habit recognition can address this problem. 

Systematic and possibly mandatory automation of connected installations would avoid waste due to lack of knowledge and negligence of users. 

The actuators enabled by the IoT can be substituted for humans to take energy saving measurement decisions (reduction of heating, switching off lights, delayed departure of household appliances). 

For users, it is also important to provide them with indicators related to their uses. The aim is to be able to show them their consumption but also the savings that they have been able to achieve with the implementation of measures.


To reduce emissions from collective buildings: sharing and pooling are key elements and AI can greatly contribute to optimising them.

Smart building is more than just for indoor use. If the building presence and usage data, passages and flows exist, it is then possible to anticipate and schedule the need for public transport. 

In the common areas, making the sharing of car parks flexible, for example between companies, but also between day and night use, makes it possible to reduce parking areas often unused in the city. 

Parking sharing will become increasingly important with the arrival of electric cars and access to shared charging terminals. If the information on the time of arrival and departure of the cars is known, then it is possible to optimise the charging flows and thus limit the power required by distributing the charge according to the planned parking times of the vehicles. 

Tomorrow, electric cars will be able to provide a real energy reservoir and they will be able to contribute to smoothing energy production to reduce the share of fossil fuels in the energy mix. 

Renewable energies are intermittent energies. Also, their production is not constant. However, one of the challenges for electricity suppliers is to be able to adapt their production to changes in consumption because electricity storage is currently a problem. Since a car is parked 95% of its time, it is possible to imagine electric cars returning power to the network when they are not used. All this will not be possible without data and the smart grid.


The challenges of the transition to smart building

The lifetime of a building is approximately 50 years, and the real estate stock is renewed by 1% per year. One of the challenges is to make the existing housing stock smart despite this low renewal rate and very short climatic deadlines.

The solution would be to make a building smart through renovation. But adding sensors and an additional layer on outdated hardware bases is not always possible. The risks associated with the reliability and relevance of data are real. And it is often necessary to initially aim for simple and fast ROI on the buildings to be renovated. 


How to delegate comfort and responsibility

There are also governance issues. Limit the temperature of a building to 18°C, when it is unoccupied or share a parking space with other users: Are individuals willing to delegate their choice of comfort to algorithms that will decide for them?

Also, sensors track us down to find out about our habits, it is a subject of individual freedom, within the framework of an imperative of collective responsibility. How can we gradually make these uses acceptable?

The effective application of sobriety will require users to be made aware of economic and environmental issues. This requires effective change management to ensure that the solutions set out above are sustainable and accepted.


Apply smart building on a larger scale to have a significant impact.

As we have seen, smart building is an effective way of controlling energy consumption. The same logic also applies on a larger scale (neighbourhood, city, country). 

A neighbourhood can adapt its lighting according to the passage and use of the public space. The city can synchronize traffic and streamline traffic using AI algorithms. A country's electricity grid can be controlled with the smart grid and the monitoring of instantaneous consumption but also the injection of energy by electric cars or solar roof panels.

All this is possible only with the proper use of the technologies we have and new technologies: IoT, 5G, Artificial Intelligence, Cloud, Big data, etc.