26 JUNE 2019 ENGINEERINGOn a journey
Parents of the revolution: why we must start planning for the coming
digital age today. A plan of action from Stephen WoodhouseD
on’t work harder;
work smarter.
This old adage is taking on
new significance as digitalisation transforms
our economies. Today, we stand at
the precipice of a global productivity
revolution and the watchword of this will
be: “work smarter.”
Yet, while this transformation will free
us from a great deal of wasted time, it
will also require us to change the energy
industry. In the future, we will all be
required to learn new skills and change
the way we work – quite fundamentally- to adapt to this emerging reality.
This requires workforces to change
their cultures and mindsets, while also
learning new ways of working. This
is no small task. Yet it is vital, because
those who succeed will find themselves
at a competitive advantage.
Digital applications in energy
have the potential to transform the
sector, by delivering greater efficiency
throughout the entire supply chain,
by revolutionising companies’
relationships with their customers, and
by unlocking the potential for deep
decarbonisation through automating
flexibility to match production patterns
of renewable energy. The earliest
digital breakthroughs are in predictive
asset maintenance, improved
forecasting and real-time monitoring,
and digital tools that aim to attract and
retain customers. Drones and UAVs
for remote inspections, as well as
process mining and text mining are also
helping to improve efficiency. Digital
twins allow ‘what-if’ and predictive
analysis to be performed on virtual
representations of physical assets.
Artificial intelligence is unlocking value
almost everywhere it is applied.
So, while this revolution will be full
of opportunity, we must ask ourselves
some tough questions: What does the
future look like? How do companies
ensure they have the right structure and
skills to lead this change? And what does
the company of the future look like?Skilling up
While still a nascent technology,
predictive asset maintenance is
becoming one of the more mature
digital technologies in the energy sector- and it tells us important things about
the changes to come. Today, predictive
maintenance is at the cutting edge, but
tomorrow it will be part of a much
bigger system. We are still at the cusp
of what the Industrial Internet of Things
(IIoT) can do.
The guiding star for all “industry 4.0”
technologies will be data. The data that
these IIoT sensors gather will enable
companies to identify and resolve
problems remotely, allow engineers
to deploy their time more efficiently
and, eventually, machine learning
might help plants automate simple
engineering jobs. It will also allow plant
owners to gain insights into their own
operations and identify how assets can
be used more productively. Energy
companies are still only at an early stage
in exploiting digital technologies and
data streams, such as machine learning
applied to rich data sources.
However, this future is not yet
here. To reach this point, we need
better access to clean, accessible
data streams and we need to better
identify where to focus our efforts.
We also need to get around practical
barriers like the interoperability of
these sensors. Although the limits
are expanding fast, constraints on
processing power, data storage and
algorithms mean that the 80:20 rule
still applies to data analytics. It is these
practical considerations that led Pöyry
to co-develop Krti 4.0, a machine
learning predictive maintenance
framework that works across different
kinds of sensors. It begins with
prioritisation using Pöyry’s RAMS
(Reliability, Availability, Maintainability,
Safety) methodology, to assess and
prioritise the most relevant potential
causes of failure. This RAMS approach
distinguishes Krti 4.0 from less
discerning systems; allowing Pöyry’s
engineering experience to direct
the system to focus on the most
important data.
Tomorrow’s barriers will be the
evolving relationship between humans
and machines: we must learn when
to trust machine decisions, how
to monitor and when to take back
control, and how to augment machine
learning with humans’ experience and
knowledge from beyond the data.
For policymakers, it may also require
changes to the way with think about
regulation, as the lines blur between
utilities and tech companies.ENERGY INDUSTRY
Stephen
Woodhouse