Where is the Edge of the Edge of Industrial IoT?
information is high, as response delays
(decision latency) of minutes and hours can
amount to significant losses. This business
problem dictates that the Edge is at the plant
area level.
Key Objective 1: Protect equipment from
damage by overheating
In this scenario, a “dumb” thermocouple can
measure temperature on a pump. A pump
with edge computing capability can perform
basic analytics to determine if a defined
threshold is exceeded. From a control
perspective, it may have the ability – in
millisecond response time – to immediately
shut the pump down. There is no decision
latency and no need for connectivity to
perform this fundamental capability. It does
not mean that it can’t be connected for
notification purposes, it is just not necessary
for this capability. The time value of the
temperature information will decay rapidly
as delayed response will result in equipment
damage. In this case the Edge will be at the
device level as it will still be able to achieve
the key objective, even if connectivity to
higher level systems and networks are
interrupted.
Key Objective 3: Optimize supply chain for
a location or factory on a twice daily basis
Optimizing supply chain processes for a local
facility, factory or an oil field requires data
from multiple sources at short intervals
(typically hours) to apply optimization
algorithms and analytics that will adapt
supply chain plans in business systems such
as SCM or ERP solutions. The fundamental
capability requires at least local or factory
level connectivity with decisions made in
hours. Additional information outside the
perimeter of the factory may be useful, but
not mandatory for effective optimization. In
this instance, the Edge is at the perimeter of
the factory, plant or local facility.
Key Objective 4: Predict equipment failure
and schedule proactive response
Key Objective 2: Proactively monitor the
performance of critical plant areas or
production lines
Building machine learning models to predict
ESP (Electric Submersible Pump) failures
requires data from multiple offshore
platforms. The analytics models are complex
and a large amount of data is needed to train
and re-train the models. It also requires
regular data feeds from operating ESPs to
determine each unit’s RUL (Remaining
Useful Life). The data from individual ESPs
need to be analyzed on a regular interval but
information decay is much slower than in the
other scenarios and decisions can be taken
on a daily or weekly basis. In this scenario,
the fundamental capability is typically
performed at the enterprise or even cloud
The performance of critical equipment and
production lines are often expressed
through performance indicators like OEE
(Overall Equipment Effectiveness). Near
real-time analytics on multiple data points
from sensors on the plant area can be
processed on a local gateway at the plant
area level and provide alerts to operational
systems or personnel on areas with specific
OEE trends, for example. In this instance, the
fundamental capability requires information
from multiple equipment sources to perform
simple analytics. The time value of
IIC Journal of Innovation
- 7 -