In the U.S. alone, buildings are responsible for almost 40% of the country's total carbon emissions, most of which come from day-to-day lighting, heating, cooling, and appliance operation—making it one of the largest single polluting factors in the country. However, improving the energy systems of these buildings is a complex, dynamic pursuit. This is an inherently chaotic space, with a mixture of control systems, interoperability, data standards, and equipment configurations, worsened by the fact that close to 65% of U.S. buildings standing today pre-date the end of the Cold War, when energy efficiency standards for building equipment (e.g., air conditioning systems) were virtually nonexistent.

"Carbon Lighthouse has a single mission: to stop climate change," said Brenden Millstein, President of Carbon Lighthouse. "Over the past decade, we've successfully been working toward this mission by developing a business model and our patented AI platform CLUES® to help existing commercial real estate buildings—one of the biggest contributors to the country's CO2 emissions—reduce their energy use and overall carbon emissions."

By using artificial intelligence and machine learning to understand usage patterns, seasonal temperature trends, and other factors that typically affect a building's heating and cooling decisions, Carbon Lighthouse is able to help commercial real estate clients save an average of 10–30% off their energy bills for a typical building with existing equipment.

"Right or wrong, capitalism is the economic engine of the world," said Millstein. "And so, in order to really make a difference within the sustainability space, it has to have a financial argument to go with it, which matters to that business, and in a timeframe that matters to their investors. And we are helping bridge that world a little bit."

Carbon Lighthouse uses advanced sensors, artificial intelligence and machine learning analytics, and data from the building and existing systems to analyze, optimize, and monitor the heating, ventilation, air conditioning (HVAC), and lighting systems—the systems using the most energy in buildings—to maximize efficiency opportunities.

"Because of the work we've done over the years in expanding our models, capabilities, and data reach, we've gone from reducing the equivalent of the emissions of a few cars 10 years ago to having reduced the emissions equivalent of 13 power plants as of the end of 2020," said Millstein.

Improving building efficiency with artificial intelligence and machine learning

To make a building more efficient, Carbon Lighthouse needs to understand how each space is used across multiple occupancy and weather variables, including things like weekdays versus weekends, a typical workday versus a holiday, or 6 a.m. in the winter versus 4 p.m. in the summer. The company needs to understand how the building systems respond in concert to these dynamic conditions. The answers lie in the data.

They approach the problem in two ways: directly tapping into building management systems and creating a building's data stream to measure things like lighting levels, temperature, relative humidity, and airflow. They can then extract these massive data streams and bring them together with the 100 million square feet of real building data already in their CLUES platform to model the energy being consumed throughout that specific building and find areas to optimize energy use. CLUES is powered by AWS, using Carbon Lighthouse-developed artificial intelligence and machine learning techniques and models that can be quickly retrained and redeployed at low cost as conditions change and new data becomes available. CLUES is continuously learning from an expanding wealth of building data it continuously ingests from each facility it services. It then passes on those learnings to the company's engineers and building's on-site property teams, who can use the information to further optimize algorithms and decrease future emissions.

"We tap the power of machine learning to analyze billions of data points—all in an effort to find hidden energy efficiency opportunities in lighting and HVAC systems and maximize carbon reductions," said Millstein.

"We turned to AWS's compute and data storage solutions to build a foundation that can continue to scale with our needs," said Millstein. "Today, Carbon Lighthouse collects 690 million new data points per building each month. With several hundred buildings under our belt (and growing), the sheer volume of data we're working with would be impossible to manage without AWS data storage solutions."

The Carbon Lighthouse platform uses a number of AWS technologies—including Amazon Elastic Compute Cloud (EC2), Amazon Simple Storage Service (S3), Amazon Relational Database Service (RDS), Amazon Dynamo DB—that offer the building blocks they need to ingest and store large volumes of data to build and deploy artificial intelligence and machine learning models, while eliminating the operational challenges that go into managing databases and worrying about data and infrastructure requirements at scale.

Their artificial intelligence and machine learning algorithms learned to model a building's energy use over time, from the building level to the system level. Carbon Lighthouse engineers can test and change model parameters to understand how the building could be operating differently and more efficiently by analyzing different scenarios.

The technology not only uncovers opportunities to dramatically reduce carbon output, but it also provides guaranteed operating expense reductions.

According to Millstein, Carbon Lighthouse's "business model and success are based on having deep insights into a building's operational systems to clearly demonstrate the financial and environmental impact of our solutions. To date, we've been able to eliminate 260,000 metric tons of CO2."