Artificial Intelligence (AI) has become one of the most talked-about topics in commercial building operations. While the term itself is relatively new to many building owners and facility managers, the underlying concepts are not. Building Management Systems (BMS), trend logs, alarm analytics and performance reporting have been identifying operational issues for decades. What has changed is the ability of modern software platforms to process large volumes of data, recognise patterns and identify anomalies much faster than a person manually reviewing reports.
In an unsupervised building environment, AI and predictive monitoring can provide an additional layer of intelligence by continuously analysing information collected from sensors, meters, equipment controllers and operational systems throughout the building. Rather than simply reacting to faults after they occur, the system can identify subtle changes in performance and alert operators before those changes become costly problems.
Pattern Recognition
Every building develops operational patterns over time.
Occupancy levels fluctuate throughout the day, HVAC systems respond to seasonal weather conditions, lifts experience predictable traffic peaks, and energy consumption follows identifiable trends. By analysing historical data, predictive monitoring platforms can establish what “normal” looks like for a particular building.
When performance deviates from these established patterns, the system can generate alerts for further investigation. This allows facility managers to focus their attention on genuine anomalies rather than manually reviewing thousands of data points.
Abnormal Occupancy Detection
People counters, access control systems, lift usage statistics and security systems can all contribute to occupancy monitoring.
For example, a retail shopping centre may typically experience increased traffic during weekends, public holidays and promotional events. If occupancy suddenly exceeds historical expectations for a particular time of day, the system can notify building operators and trigger additional responses such as:
- Increased cleaning schedules
- Additional security patrols
- Escalator optimisation
- Lift performance adjustments
- HVAC capacity increases
- Waste management responses
Conversely, unexpectedly low occupancy may indicate access issues, tenant closures or other operational concerns that warrant investigation.
Chiller Efficiency Drift
One of the most valuable applications of predictive monitoring is identifying equipment performance drift.
A chiller may continue operating and maintaining building temperatures while gradually becoming less efficient. Without detailed analysis, this issue may remain unnoticed for months or even years.
By comparing:
- Power consumption
- Cooling output
- Condenser water temperatures
- Chilled water temperatures
- Outside air conditions
- Historical performance data
the system can identify efficiency losses before they become significant operational or maintenance problems.
This allows maintenance to be scheduled proactively, often reducing energy consumption and extending equipment life.
Water Consumption Anomalies
Water metering and trend analysis can identify unusual consumption patterns that may indicate leaks, faulty equipment or operational inefficiencies.
Examples include:
- Cooling tower make-up water increases
- Domestic water leaks
- Irrigation system faults
- Toilet cistern failures
- Hydrant or fire service leaks
- Tenant water consumption abnormalities
Rather than discovering the issue through a large water bill weeks later, predictive monitoring can identify the change shortly after it occurs and notify the appropriate contractor or facility manager.
Energy Spikes and Abnormal Consumption
Energy meters installed throughout the building provide a detailed understanding of where electricity is being consumed.
Predictive monitoring platforms can analyse:
- Building demand profiles
- After-hours energy consumption
- Tenant energy usage
- HVAC energy performance
- Lighting energy consumption
- Equipment start-up characteristics
The system can then identify unusual spikes or gradual increases in consumption that may indicate operational problems.
For example, a supply fan running continuously after hours, a failed sensor causing excessive cooling demand, or a lighting control fault may be detected long before a monthly utility account is received.
Contractor Performance Tracking
The same technology used to monitor building systems can also be applied to contractor performance.
Through QR code attendance systems, work order management platforms, access control records and maintenance logs, building owners can gain greater visibility of contractor activity.
Metrics may include:
- Response times
- Attendance compliance
- Time on site
- Work completion rates
- Repeat fault attendance
- Preventative maintenance compliance
- Service level agreement performance
Over time, this information creates a measurable performance profile for each contractor, helping facility managers make informed decisions regarding future service agreements and contractor selection.
From Reactive to Predictive Operations
Traditionally, commercial buildings have operated in a reactive manner. Equipment fails, alarms occur, tenants complain and contractors are dispatched.
Predictive monitoring changes this approach by identifying trends before they become failures.
For unsupervised buildings, this additional layer of intelligence can significantly improve operational visibility while reducing unnecessary callouts, minimising downtime and improving asset performance. It enables facility managers to focus on decision-making rather than data collection and provides landlords with greater confidence that their assets are operating efficiently and reliably.
Importantly, AI does not replace facility managers, engineers or contractors. It simply helps them identify opportunities and risks sooner. Human expertise remains essential for interpreting information, assessing priorities and making operational decisions. The technology acts as another tool within the building management toolkit, helping people manage increasingly complex buildings more effectively.