Buildings account for 38% of global carbon emissions. It stands to reason, then, that improving their energy efficiency will make a major impact in the fight against climate change.
In fact, we know this is true because we’re already seeing it happen. Over the last decade, the population of Greater London grew by over 7%, and pre-Covid the city saw a 140% year-on-year increase of buildings over 19 storeys. Despite this, between 2000 and 2018, London achieved a 57 per cent reduction in workplace greenhouse gas emissions. According to the Institute for Market Transformation, this is largely thanks to the decreasing energy use intensity in commercial buildings.
Measure to cut
So how do we follow suit and reduce energy use in our own buildings? This is a question many building owners are asking. As the saying goes, if you can’t measure it, you can’t manage it. Therefore, the first step is to find out exactly how much energy is being consumed.
Regulations such as Minimum Energy Efficiency Standards (MEES) work in tandem with Energy Performance Certificates (EPC) to compel owners and facility managers to achieve greater energy efficiency, thereby lowering their buildings’ emissions. To do this efficiently, people are turning to AI and cloud computing tools to record, manage, and visualize their data.
The second step is to find the main energy thieving culprits in the building. Thanks to an increasing body of data and research, we know that a commercial building’s heating, ventilation, and air conditioning (HVAC) system, accounts for up to 50% of a building’s energy consumption. Much of this energy isn’t even going to good use – in fact, up to 30% of it is wasted due to inefficient, outdated equipment.
AI to heat and cool
One might think that the solution would be to completely overhaul our HVAC systems. While retrofitting is certainly one way of going about it, the less expensive, less invasive, and faster alternative is to integrate AI into existing systems, effectively rendering them intelligent.
Of course, this sounds simpler than it is. You can’t just plug AI into your HVAC system and hope for the best. A good AI solution needs the right parameters. That’s where prior institutional knowledge and data come in. It’s important to have a firm initial grasp of how an HVAC system works. After that, you can leverage a plethora of data available from building sensors and equipment and layer on top of that external sources such as weather forecasts, occupancy rates, utility rate providers, and live carbon indices on the electrical grid. These can all work together to create the optimal solution tailored for each unique building.
But perhaps the greatest aspect of using AI models to optimize HVAC systems is that they can be trained continually as new information is collected, adapting to changes in temperature, occupation, or any other relevant information. In short, they are constantly auto-tweaking into more efficient versions of themselves. Since an AI model has the capacity to communicate directly with a building’s central control system, or building management system, intervention wouldn’t be necessary. Instead, left to its own devices, it uses time-series prediction models to accurately predict temperature changes and autonomously adjust accordingly, becoming more effective over time.
The benefit of this kind of technology extends beyond a reduction in emissions. It means a decrease in costs, an increase in occupant comfort, and a lengthening of HVAC equipment life. Viewed holistically, it certainly doesn’t take a giant leap of the imagination to deduce that AI could really revolutionize the corporate real estate industry while making a big dent in GHGs.
Scale to impact
But to make the kind of impact needed to turn the tide on the climate emergency, this technology needs to be rapidly scaled and adopted on a global level. According to the Capgemini Research Institute, this is entirely doable. In fact, it projects that by 2030, AI is likely to have reduced overall GHG emissions by 16% and will have helped organizations fulfill up to 45% of the Paris Agreement targets.
The reason AI has such prolific potential is that it can be adapted and employed on a sizeable scale, collecting masses of data and using it to significantly lighten the load on a city’s electrical grid. In a world where extreme temperatures, rising populations, increasing electrification, and weather-dependent renewables are putting unprecedented strain on our outdated grid systems, we are in desperate need of a stabilizing force that encourages greater grid flexibility.
And AI presents a compelling answer. After all, if AI could reduce a single building’s carbon footprint by up to 40%, imagine what it could do with whole portfolios of buildings connected by a collective model that uses cross-sectional data to cooperatively reduce energy consumption. Consider the consequent effect this could have on entire cities and their grid systems. If large constellations of buildings were optimized and fine-tuned according to their own individual needs in conjunction with the needs of the grid, avoiding peak times when power is dirtier and costlier, they could become active participants in day-to-day grid operations. This could support both a more economically and environmentally efficient approach to energy consumption while creating substantially more sustainable cities.
While AI is by no means a hard and fast answer to climate change, its scalability, accuracy, predictive ability, and potential to process massive amounts of data make it one of the most impactful tools we have to reduce carbon emissions in both individual buildings and, perhaps more powerfully, in larger portfolios of connected real estate. This technology is not an expensive pipe dream. It’s cost-effective, it works, and it already exists. All we need now is its adoption.
