Carbon Lighthouse is using analytics powered by AWS to fight climate change.
Operating commercial, industrial and residential properties in the United States accounts for more carbon emissions than transportation, according to the Environmental Protection Agency. Transportation accounts for only 29% of carbon emissions, while electricity, industry, commercial and residential property, and agriculture combine for 68% of carbon emissions.
Carbon Lighthouse, therefore, made it its mission to reduce carbon emissions produced by commercial and industrial buildings to help thwart climate change.
The organization, founded in 2009 and based in San Francisco, aims to accomplish its mission by offering energy savings as a service to make it profitable for building owners to reduce their energy needs and the carbon emissions that result from those needs.
To date, Carbon Lighthouse has reduced more than 260,000 metric tons of emissions, which is equivalent to the energy produced by 18 power plants, while at the same time delivering $250 million in savings for its clients, according to Brenden Millstein, president and head of product at Carbon Lighthouse.
At the core of the organization's battle against climate change -- its ability to help clients reduce carbon emissions, while at the same time delivering cost savings -- is analytics.
"Carbon Lighthouse makes it profitable for commercial and building owners to decarbonize their portfolios," Millstein said in a breakout session during AWS' Machine Learning Summit, a virtual user conference.
Carbon Lighthouse developed a software platform called CLUES (Carbon Lighthouse Unified Engineering System) that analyzes more than 100 million square feet of clients' commercial and industrial real estate and 5 billion energy data points. CLUES is an augmented intelligence platform that collects building data in real time and uses machine learning to develop models that lead to insights about how to reduce carbon emissions and what the resulting savings will be.
Underpinning CLUES is a technology stack from AWS.
Carbon Lighthouse uses Amazon Elastic Compute Cloud to run its servers, Amazon Relational Database Service to store metadata such as building addresses and square footage, Amazon Simple Storage Service to capture and store time-series data such as the data collected on site at buildings, and Amazon DynamoDB to retrieve and work with data.
Brenden MillsteinPresident and head of product, Carbon Lighthouse
"It all starts with data that is streaming into our cloud platform from the 100 million square feet we already have in the platform," Millstein said.
When a new client signs a portfolio with Carbon Lighthouse, the organization sends sensor kits to them and starts new data streaming from their buildings, Millstein explained.
The recommendations generated are not simple, such as turning off the lights or an air conditioner at night. They're more complex -- for example, directing control systems so specific valves are automatically opened or closed at a specific time or when the temperature in a building reaches a specific level.
It's about continuous optimization to reduce the effects of climate change using analytics, according to Millstein.
"It's not major changes -- that's the exciting thing," he said. "Don't think replacing the windows; don't think cranes and boilers. This is mostly making the existing equipment slightly better. For clients, it saves a lot of time and results in real CO2 savings, which is the whole point."
One of the organizations Carbon Lighthouse works with owns a building near downtown Los Angeles on Wilshire Boulevard.
The 400,000 square foot office building was already energy efficient. Its heating and cooling system is similar to most heating and cooling systems, with air handlers that have big fans and coils of water. When the building needs cooling, the fans blow air over coils with cool water, and when it needs heating, the fans blow air over coils with hot water.
Carbon Lighthouse provided the building with sensor kits to capture data, and the data captured from the building's 70 air handlers -- plus boilers and chillers in the basement and cooling towers on the roof -- was then used to inform the control system.
Rather than turning on all 70 air handlers at once, the control system now has the intelligence to know how many need to be turned on and how many can be turned off to use the minimum amount of energy needed to maintain a comfortable temperature.
"It's this constant tweaking and tuning to gain maximum performance out of the building," Millstein said. "This is a big deal."
The result, he continued, was $200,000 in annual savings, which was about 25% of the building's energy costs. That annual savings, meanwhile, increased the value of the building by $3.5 million, Millstein said.
And when the owner of the building on Wilshire Boulevard later installed Carbon Lighthouse's technology in a building in San Francisco, Millstein said it led to the combined elimination of 7,900 tons of carbon dioxide, which is equal to the amount of carbon emissions produced by driving a car 17.7 million miles.
But just as Carbon Lighthouse is helping clients become more energy efficient using analytics, the organization itself is becoming more efficient as it fights climate change. Forced by the COVID-19 pandemic to make adjustments to its business model and become more efficient, Carbon Lighthouse was able to eliminate more carbon emissions during the first quarter of 2021 than it did in all of 2020, Millstein said.
"It's an exciting time for the planet," he said. "And an exciting time for engineering."