Accelerating Performance with the Artificial Intelligence of Things
Smart electric grids have already
proven more efficient in connecting
renewable resources, improving
system reliability and billing
customers on more granular usage
increments.
Machine-monitoring sensors already
diagnose and predict impending
maintenance
issues,
trigger
deliveries where and when needed,
and
prioritize
maintenance
schedules.
Data-driven systems are being built
into the infrastructure of “smart
cities,” making it easier for
municipalities
to
run
waste
management, law enforcement and
other programs more efficiently.
Whatever the industry, there are use cases
in place to learn from and build on.
A report from the McKinsey Global Institute
estimates that the IoT could have an annual
economic impact of $3.9 trillion to $11.1
trillion by 2025 across many different
settings, including factories, cities, retail
environments and the human body. 13 With
AI, the realization of this impact is getting
closer everyday – and first movers take the
prize.
Leaders should continue to innovate in areas
such as:
13
without explicitly being told where to
look or what to conclude, resulting in
better, faster discoveries and action.
Natural language processing (NLP)
to enable machines to intelligently
interact with humans, such as via
chatbots, and discover insights in
large amounts of digitized spoken
content.
Computer vision to analyze and
interpret what’s in a picture or video
through image processing, image
recognition and object detection to
provide insight not easily available
with existing sensors. Cameras can
sometimes be installed much more
easily than other sensors. Cameras
can also be retrofitted to existing
assets passively, where adding
sensors can more invasive. This is the
technology behind the SciSports
success.
Forecasting and optimization to help
AI systems predict future outcomes
based on IoT data and deliver the
best results under given resource
constraints.
Recurrent neural networks (RNN) to
use sensor data to frequently capture
measurement over time and look for
issues that develop. RNNs can
provide complex pattern recognition
as well as specialized forecasting.
Machine learning and deep learning
to find insights hidden in IoT data
Patel, Mark. Shangkuan, Jason. Thomas, Christopher. What’s New with the Internet of Things? McKinsey & Company.
IIC Journal of Innovation
- 48 -