• The solution, currently being tested at various Orange facilities in France, has now been deployed in 15,000 buildings across 70 cities worldwide.
• The company has also developed ARIA, a virtual assistant that can provide building managers with predictive analyses and maintenance recommendations. Brainbox AI sales engineering professional Jean-Michel Provost-Théroux explains how it works.
BrainBox AI was one of the exhibitors at Orange OpenTech in November. Can you tell us about your company?
BrainBox AI is a Canadian company with 200 employees. We were spotted by Orange in 2022 and subsequently given support by the Open Innovation team, the group’s start-up accelerator. That was when we realized that Orange’s sustainable development goals were very much aligned with our plans to make buildings smarter while reducing their carbon footprint.
We have developed an AI system that is trained on data from building management systems, which have sensors for humidity, CO2, and temperature and so on. This data is transferred to the cloud and processed by the AI which uses it to generate intelligent commands, which are then returned to individual buildings where they optimize operating parameters so as to reduce energy consumption and carbon emissions. For enhanced accuracy, the AI, it also takes into account meteorological and power grid data in addition to information sourced from buildings.
The long-term objective is to provide professionals with an AI assistant that will help to identify and resolve problems in just a few minutes
What are the main features of ARIA and how does it help managers to rapidly resolve building maintenance problems?
Artificial intelligence provides comprehensive software solutions, but buildings still have problems that require physical intervention. That is why we developed ARIA, a virtual building engineer, which acts like the AI Jarvis [a reference to the Marvel films, editor’s note] to assist building operators and managers. It is a generative AI trained on building data that provides support to building managers, operators and maintenance staff with advice and analyses and also predictive maintenance recommendations to ensure proper upkeep. The long-term goal is to provide these professionals with an artificial intelligence agent that can identify and resolve problems in just a few minutes, offering revolutionary gains in productivity to the industry.
Can you give us some concrete examples of problems that ARIA can identify and solve in buildings?
For example, you can ask ARIA, either by text or voice, about the type of equipment present at a particular site and the type of maintenance that should be carried out there. If an operator receives a complaint about a floor where the temperature is too cold, ARIA will be able to analyse data on the building to generate leads that help identify the source of the problem, which might be caused by a defective valve in a heat exchanger, for example. The goal is to optimize the continual maintenance of buildings to ensure that they consume as little energy as possible. For their part, infrastructure managers, can ask ARIA for detailed analyses of energy consumption over given periods, for example during extreme cold spells and heatwaves, so as to optimize thermal insulation and air conditioning…
Can you tell us about the results you have achieved since the launch of BrainBox AI and your objectives for the coming years?
BrainBox AI is now deployed in more than 15,000 buildings in several countries. In France, we are setting up partnerships with different entities in Orange Group. In the United States, we are already deployed in 60 major hypermarkets. Since our launch in 2019, our solutions have been adopted in 70 cities worldwide. These are encouraging results: with BrainBox AI we can cut total HVAC (heating, ventilation and air conditioning) energy costs in buildings by up to 25%, which in turn cuts their emissions by up to 40%, and at the same time provide a 60% improvement in occupant comfort.
Jean-Michel Provost-Théroux
