Julian Rimmer of M Moser Associates shares how they are using AI and machine learning to help clients realize their corporate real estate and business goals.
The last few years have seen a dramatic change in the types of technology used in the built environment’s design, build and operation. The most exciting and flexible of these tools, machine learning (ML) and artificial intelligence (AI), are becoming increasingly sophisticated and broader in application every year. Here’s how we are deploying these tools across our services to help clients realize their corporate real estate and business goals.
AI in the workplace
AI tools are highly touted to solve most challenges today with a broad range of capabilities and functions. In the built environment, these tools are helpful only to the extent the building, systems and corporate policies allow them to be. There are many areas where AI can be deployed in the workplace, from sensors and analytics to automation, communication and reporting. Their success heavily depends on a client’s corporate real estate strategy, culture and technological ambition.
A common use of AI in the workplace is to collate user feedback, such as comfort surveys, thermostat and light level adjustments, and other data, such as weather, electrical grid carbon intensity and space occupancy, to develop an optimal operating schedule over time. This way, the AI tool learns how the workspace operates and slowly adapts conditions proactively to optimize comfort and energy use.
AI in M Moser’s living labs
Our living lab program offers real-world environments to test these tools independently and together as part of a system. One problem we are trying to solve is that the built environment is intricate and full of inertia. Legacy technologies from past projects often remain and need careful consideration. This means that recent technologies need to be tested for their utility and ability to integrate into a smart building platform, as well as how they support a multi-site portfolio.
Our smart building program spans 12 global offices, all connected to a portfolio cloud service. This connection facilitates continuous testing, ensuring that when we integrate these technologies into a project, we are confident in their performance and compatibility with the overarching smart building platform.
One example of this is a validation project we undertook with a supplier of smart sensors. The project was intended to confirm their current sensing technology, which draws from various environmental factors and then uses AI to provide an accurate picture of occupancy and thermal comfort. The project brief was extended to jointly push the sensor’s capability to include people counting using AI. In partnership with the vendor, we identified the requirements, where and how it should be used, and the level of accuracy it could provide. We then used actual people counting to verify the accuracy of the sensor’s algorithm. This AI-enabled function has been rolled out on several of our projects today.
How AI helps occupiers
AI is a powerful tool for clients in delivering occupant experience, monitoring space performance and maintaining equipment.
We are supporting a global banking client to implement a smart building system at its headquarters in New York City. This project will feature one of the broadest uses of technology and AI in a commercial office globally. Central to the client’s design brief is for the project to reflect their diversity, equality and inclusion (DEI) values across the design. Working with our DEI and workplace strategy teams, our digital buildings team has deployed sensors capable of reporting environmental conditions to a digital twin, essentially a cloud-based copy of the building operating data. The digital twin then analyses and uses AI to serve real-time and predicted statistics back to user engagement dashboards and mobile applications. This allows occupants to find a working environment that meets their needs.
This system also pulls data from a range of AI-enabled sensors and then uses additional cloud-based AI tools to help their facilities team identify equipment which may need servicing. Based on equipment energy consumption and vibration, this predictive maintenance method enables the facilities management team to focus service calls on equipment that needs servicing rather than based on a schedule. One of our clients, a global entertainment company, has seen a reduction in service cost by over 50% by employing these tools in its London headquarters.
It isn’t about how little space you can get away with, it is about how to best use the space you have available to drive business forward.
Last year, we worked with a global financial services client to pull real-time energy data from its London office and real-time electrical grid carbon intensity data from a web service. The system then can automatically generate compliance and sustainability reports for operational carbon. It can also predict operational carbon intensity to shift power-intensive processes to periods where it is more efficient.
One of the most important use cases for AI tools in the workplace is in evaluating space performance and the alignment between the office’s needs and what can be delivered by the workplace. This allows corporate real estate (CRE) executives to make better decisions about how to deploy space. AI tools use space data to form a picture of what is happening and where. Crucially, while these AI tools are powerful, they don’t answer the “why,” which is where human intervention and observation play a crucial role in the technology to provide insight to clients.
We are doing this across several sites now to help clients make the best decisions possible. It isn’t about how little space you can get away with, it is about how to best use the space you have available to drive business forward.
Benefit of AI on employees in the workspace
The relationship between occupants and technology is a complex subject and a fine balance needs to be struck to provide valuable, people-centric services that are not intrusive. Outcomes from AI tools, such as the dashboard described above, can be compelling in supporting change, enabling transparency and communicating values – such as sharing decarbonization progress with employees. This is especially poignant today as work patterns and expectations continue to shift in response to the climate crisis and post pandemic.
The data-rich environment also supports DEI initiatives, providing on-demand information and trends that can help support neurodiversity and accessibility needs. Solving complex control issues, such as balancing increased ventilation for health with energy savings for decarbonization, is beneficial.
…The highest mark AI can score in the workplace is that it goes unnoticed.
Tesla Motors has a design mantra – “All input is error”. This principle is beneficial when developing a data and AI strategy as it essentially means that the system should be able to operate in an optimal condition at all times. If a human needs to continuously override the system – by adding input, it has missed something. In the built environment, this could be regularly adjusting a thermostat, changing a blind setting, or dimming lights. To this end, when we successfully implement AI across sensors, services, building controls and automation, the occupant should be entirely unaware of its existence. The space should always provide the right conditions. Ironically, the highest mark AI can score in the workplace is that it goes unnoticed.