The data extracted from these various systems
is streamed over the Internet and into the cloud.
In the cloud, the data is then structured and
aggregated for the purpose of data analytics.
Data in the cloud is analysed using the various
algorithms available; to produce business and
operational insights. Some of the more complex
algorithms include anomaly detection and machine
learning. Anomaly detection studies the data for
unusual events not consistent with the streaming
data. The detected fault is then investigated to
identify the root cause for immediate rectification.
Machine learning on the other hand, analyses
historical data to predict possible future outcomes.
Thus, potential failures in systems can be detected
beforehand and avoided altogether. Real-time
streaming and analytics of data enables the
reduction of time-based preventive maintenance
work and encourages more condition-based and
predictive maintenance work.
For a typical building, the analysis of utility
consumption data helps to eliminate wastage
and leakages in energy and water use while the
analytics of lift usage trends and washroom usage
patterns provides a rare insight into the behaviour
of the building population and their movements.
Such information provides the building manager
with a better understanding of the building’s
demands and enables him to operate the building
in a more efficient and effective way.
One of the important aspects of analytics is
the visualisation of the data and insights on a
dashboard.
The dashboard is accessible on a computer,
tablet or mobile device and it can be accessed
anytime, from anywhere in the world over an
internet connection. Its capabilities are not limited
to monitoring only, but includes control as well,
depending on the authorisation level.
13