Industrial Intelligence: AI’s Implications on Security, Seamlessness and Services for the IIoT
particular steady-state temperature. If the
Supervisor detects than an 80% duty cycle
results in a lower temperature, an alert is
sent to a user to check whether the heating
element or sensor have failed, or if an
insulating door was left open. If an extruder
notices that the processing force increases
suddenly, the system can identify that the
die has yielded.
lives, blending into the background of
industrial environments.
AI can also be used to reduce user friction,
integrating smart objects transparently into
conventional workflows or allowing them to
operate on their own in the background.
Voice
assistants
provide
pervasive
intelligence by utilizing AI to allow
individuals to speak to a computer as they
would a person, while other technologies
leverage AI to proactively control
environments based on learned trends and
current conditions. In both cases, the ease of
interaction allows these smart devices and
systems to fade transparently into the
background, while they simultaneously
make worker’s lives more comfortable and
optimize processes.
Using our IoT-as-a-child example from
earlier, AI can teach devices and services
right from wrong, not to believe everything
they hear, not to take advice from strangers,
and cause and effect. These examples show
the potential for Artificial Intelligence to
sup ervise and protect digitally mirrored
systems .
Seamlessness
These same machines may even learn about
their environment in order to allow for
automated process optimization at scale,
reducing or eliminating the need for human
involvement. IoT has changed how we see
machines, but it also changes how machines
see themselves – with richer data and
Artificial Intelligence, machines are now able
to learn their context and to understand
their place, purpose, and processes.
Machines that understand their role in an
organization may optimize at scale and in
aggregate as opposed to individually. These
devices can then reduce a factory’s
operating costs, energy use, work-in-
progress, and exposure to safety hazards
with a minimum of human interaction. This
invisibility paired with efficacy at scale is
what makes AI such an integral, and non-
invasive, part of our daily lives.
Artificial Intelligence allows IoT to become
seamless in industrial environments.
Intelligence helps devices find pervasive
deployment, creates frictionless user
interactions and intuitive integration into
conventional workflows, and simplifies
human hand-offs.
IoT works best when it is implemented at
scale, though resource constraints challenge
IoT’s deployment. Using the same models
described in the previous section, AI can be
used as the basis of “Data Proxies” capable
of turning sparse input signals into rich
digital representation of physical objects. 3
Reducing the amount of data required to
mirror a system using Artificial Intelligence
lowers costs and allows the pervasive
deployment of devices. With smart,
connected things everywhere, they can
become an integral and invisible part of our
3
The Future Internet of Things: Secure, Efficient, and Model-Based http://ieeexplore.ieee.org/document/8055694/
- 24 -
November 2017