Demand for smart buildings is growing: For example, the market for Internet of Things (IoT) platforms is set to reach US$101 billion by 2030.
This growing appetite for smart solutions is understandable, given the operational pressures building managers are inundated with daily. Manually processing vast amounts of data, acting accordingly, ensuring occupancy comfort, and keeping energy costs and consumption as low as possible is essentially unfeasible. Overstretched building management teams mean more errors occur, and human energy is ultimately wasted on repetitive tasks.
The buildings of tomorrow are inherently self-sufficient. Data collated from IoT devices, sensors, meters, and cameras is synthesized and analyzed by AI in real time. This facilitates automated adjustments, like reducing air conditioning on an empty floor. AI can also help guide management teams on pre-scheduling maintenance ahead of a breakdown or flagging a potentially unnoticed security anomaly.
AI is revolutionizing building management. However, it can only do so with the right groundwork in place, which includes clean data, interoperability with existing systems, an understanding of AI, and a grasp of data management.
Don’t Fall for the Automation Trap
AI is often confused with automation, when the two are fundamentally different. There are widespread misconceptions in building management that AI simply operates as a “black box” and automates processes in mysterious ways.
The automation trap refers to automation that is falsely marketed as AI, leading to a patchwork of disconnected technologies operating within a system. That disconnection between tools is dangerous, especially when building managers are misled to apply these wrongly packaged technologies to smart HVAC systems that break down and cost thousands of dollars in unnecessary repairs.
AI is not synonymous with automation but can facilitate it. The distinguishing feature of AI is that it is programmed to think and learn like humans from the data it’s fed with, adapts accordingly, and can help with decision-making because of that innate learning ability. By understanding this crucial distinction, building managers are better equipped to adopt AI tools.
Prioritize Strong Data Management
Data determines how effective and reliable AI tools are for building management. Error-riddled data means faulty outputs. For example, AI used to inform smart HVAC management is supplied with inaccurate room temperature data from faulty sensors, which report a number of rooms on a floor as hotter than they actually are. As a result, the AI indicates that these rooms need cooling, wasting energy, racking up higher bills, and undermining occupant comfort.
This is a very common problem. Poor data management is one of the biggest hurdles to reliably and efficiently deploying AI. Data siloes and fragmentation worsen disconnects between technologies, undermining a building’s digital infrastructure.
Good data management is non-negotiable, especially for building managers looking to adopt systems like IoT technologies. This only means a massive influx of more data to handle, and this challenge snowballs as new technology and tools are added, especially for teams overseeing more than one building.
The building blocks of good data management include high-quality data, a clear governance framework, security, and interoperability.
- High-quality data is complete, accurate, consistent, and up to date.
- Data must have security protocols in place to protect it, prevent leaks or cyberattacks, and ensure accountability for who accesses it and where.
- For smarter analysis, data must be interoperable and integrated within a unified platform, especially when dealing with multiple systems like BMS and IoT systems.
AI needs a strong data backbone to function optimally. When data is correctly managed, building managers can maximize the benefits and use cases of AI, whether that’s to help automate temperature controls via AI, gauge insights and opportunities for reducing energy consumption, or keep a pulse on maintenance needs to minimize costs.
Train Teams to Oversee AI
The ultimate goal with AI is to have it smoothly operate in the background, taking the weight of manual legwork off building management teams. It should also be quietly ensuring that facilities are carefully monitored, data and key insights are immediately collected and processed, and alerts are issued when needed.
However, AI is not here to replace humans. People need to be kept in the loop, even if only a light human touch is needed most of the time. As with any technology, teams must have a contingency plan in place.
Protocols should be set in advance so teams know what steps to undertake in unexpected situations, like cybersecurity attacks or system failures. There is a range of situations that building managers should prepare for when adopting AI, including:
- Anomalies in outputs that can impact decision-making.
- Approvals for drastic courses of action like system shutdowns.
- Manual overrides in case of emergency or security breaches.
- Pre-defined emergency procedures in case of fire or other safety threats.
- Auditing and reviewing AI performance, including how it reaches decisions and uses data.
- Monitoring data health and monitoring for anomalies that threaten the overall reliability of the AI tools.
People need training and innate insight to feel empowered in overseeing AI and knowing when they need to step in. This also helps allocate human resources more effectively; when there’s no need to step in and AI is performing reliably, building managers can confidently focus on high-value strategic tasks.
Besides familiarizing themselves with the fundamental differences between AI and automation, individuals should have hands-on training to be aware of AI use cases in building management. Demo workshops and courses are excellent forms of hands-on training, where employees can learn to navigate dashboards and interpret insights. They also need to understand how to follow protocols and which steps to take in certain situations.
AI has the potential to be the “perfect intern,” revolutionizing building management for a smarter future. To make that a reality, the teams behind the technology need to be deeply familiar with its fundamental characteristics, ensure excellent data management, and be well-versed in working alongside AI tools.
