The Infosys Utilities Blog seeks to discuss and answer the industry’s burning Smart Grid questions through the commentary of the industry’s leading Smart Grid and Sustainability experts. This blogging community offers a rich source of fresh new ideas on the planning, design and implementation of solutions for the utility industry of tomorrow.

« 10 key pointers for an effective Web-GIS implementation leveraging ArcGIS Server | Main | The Security trap »

The Asset Management Journey - into Adaptive

For utilities, traditionally most asset management was based on cycles of planned maintenance, interrupted by many occurrences of reactive work. The planned maintenance was generally based historic norms, often with little feedback of benefit. With the advent of asset management systems, both IT (e.g. EAM/WAM) and Process (e.g. PAS55, now ISO 55000), work became more planned, and was more based on benefit, drawing particularly on asset risk and criticality. Such changes made major improvements in efficiency, with reductions of reactive work from 70% to 30% not uncommon. However planned work was, and in many cases still is, based on expectations of asset lifecycle performance, and not on actual asset feedback. Whilst such proactive strategies reduced service impacts, it led to higher levels of planned maintenance than necessary to ensure optimum asset life.


Over the last 20 years industries have increasingly turned to predictive methodologies, using sensors and instrumentation, coupled with appropriate analytic software, to predict and prevent asset failure though understanding trends. For example, a large transmission operator uses transformer load measured against ambient and internal temperature. A band range of 'normal' internal temperature against load and ambient temperature is mapped, and the system flags when internal temperature is outside of this range, so that checks can be made before any failure. Increasingly such tools are using machine learning which further helps to predict 'normal' asset behaviour. Asset management has therefore moved from Reactive through Proactive to Predictive.


Artificial Intelligence (AI) tools, such as Infosys NIA, are now starting to be used in asset management. These new methodologies use the AI engine to collate, compare, analyse, and highlight risks and opportunities. The tools can use structured and unstructured data, static and real time, and have the ability to take data from disparate sources. The systems will increasingly refine understanding of asset behaviour based on multiple inputs, such as sensors/instrumentation, third party data (weather), social media feeds, and impacts flagged by external, but publically available, sources. The tools will then be able to advise courses of action based on current events. They could also then be used to model possible scenarios, and advise actions and impacts based on their understanding of inputs against outputs (stochastic modelling +). Such tools will enable an organisation to continuously adapt its asset management strategies and implementation to current and future events.


I call this Adaptive Asset Management.

Post a comment

(If you haven't left a comment here before, you may need to be approved by the site owner before your comment will appear. Until then, it won't appear on the entry. Thanks for waiting.)

Please key in the two words you see in the box to validate your identity as an authentic user and reduce spam.