IFS logo

INTERVIEW

Artificial Intelligence’s Role in the Future Utility

Carol Johnston, VP, Energy, Utilities & Resources, IFS

1. AI has exploded onto the utilities scene recently. To what extent do we need AI to achieve a green energy transition?

The scope and timeline for the transition needed and the complexity of managing the future energy grid– with billions of distributed energy resources and a two-way flow– is absolutely reliant on a corresponding digital transformation…and AI in particular is shaping up to be a real game changer.

Enabling better insights and decision-making support, but also filling in the gaps in resource availability and experience that has been fuelled by boomers ageing out of the workforce and difficulties in attracting younger people to join the sector.

2. Powering AI is already pushing electricity grids to the limit, especially with the increased number of data centres. How can this be managed?

Yes, a double edged, or even triple edged, sword isn’t it?

We’re seeing the demand curve spike and shift from electrification initiatives and demographic growth…but for sure data centres with their geographically concentrated load profile and 24/7 operations.

Great for revenue when you’re in the business of selling power, and of course utilities are becoming bigger and bigger consumers of these data centres themselves. But how do you build out a grid able to handle these kinds of load, without blowing the roof off affordability metrics?

One way is through careful planning and partnership with communities and other stakeholders. That way, when data centres are planned and come online, renewables can be incorporated into the construction to offload some, or all, of the load during peak periods when the grid is under pressure.

And frankly, once we solve this load issue, commercial EV fleet charging is the next big mountain to climb. Where can these stations be located, and who should foot the bill for building out the required capacity to handle this load?Enabling better insights and decision-making support, but also filling in the gaps in resource availability and experience that has been fuelled by boomers ageing out of the workforce and difficulties in attracting younger people to join the sector.

3. In terms of the utilities space, where do you see AI causing the most disruption?

Well AI isn’t new, while it’s certainly gotten more attention and traction in the last few years. It’s been around for 80 + years now and since its first introduction, we’re seeing AI being leveraged and envisioned in the utilities sector to assist with business and grid analysis and planning, automating and improving customer service all the way to advancing asset maintenance and field service activities.

Imagine being able to analyze, predict and respond to asset failures…before they happen! That’s no longer fiction but a reality. Or having a system recommend crew compositions and deployments to match forecasted workloads.

The opportunities it seems are quite literally endless. At IFS we’re focused on innovating around 6 key AI themes and utilities use cases:

At IFS we’re focused on innovating around 6 key AI themes and utilities use cases:

1) Forecasting & Simulation
•   Being able to forecast event impacts to damage to critical assets
•   Forecast and analyze demand by type, location, priority, seasonality, etc and model required capacity
•   Develop inventory & equipment requirements against planned work schedule

2) Optimization
•   This one has been around for a while now, and was one of the first AI based use cases for utilities to schedule and route optimize work orders…but this technology continues to evolve and now supports dynamic scheduling, considering EV charging stops in route and much more
•   Route planning in support of carbon offset calculations
•   Automatic work bunding at the same location
•   Restoration activities after a major event, considering critical customers, number of customers impacted and other criteria

3) Anomaly Detection
•   Lots of use cases around real-time asset health monitoring & analysis
•   Asset predictive or preventative maintenance
•   And vehicle use monitoring

4) Recommendations
•   Having the system make recommendations for actions to be taken based on risk-based issue detection
•   Workforce deployment & schedule recommendations
•   Crew matrix recommendations

5) Contextual Knowledge
•   Helping users easily and intuitively navigate our software with AI powered contextual user guides
•   Similarly, helping users complete quick and accurate data capture and ensuring they working efficiently and safely by following company business processes with smart guides

6) Content Generation
•   Automating capacity plan adjustments needed to meet demand
•  Offering personalized training materials
•   And finally, for now, automating purchase & work order creation based on certain triggers

All with the end goal in mind to improve customer experience, operational efficiency and open up new revenue streams for our customers.

4. What are the risks to deploying AI, what are the risks of not deploying AI?

With any new technology there is the risk of unknown– or unexpected– consequences of adopting too quickly, or without sufficient guardrails in place.
I think we all know and have plenty of examples of how critical the grid is to our daily lives and the economy. This isn’t an industry where taking big risks without safeguards in place are rewarded. In fact, it often comes with big penalties.

With AI in particular, while it shows great promise…the recent dive in stock prices not withstanding…to dramatically improve utilities operations, we need to ensure it is deployed ethically.

You may have heard of terms like unconscious bias, or hallucinations, where AI models are poorly built, monitored and managed and give us bad results. Essentially doing more harm than good. Maybe not to the level of a Terminator level event, but not good.

But turning away from all risk and deciding not to embrace emerging technologies like AI, machine learning or even simple automation will see organizations left behind. AI fuels innovation, operational efficiency, new business models, strategic decision making and unlocks new revenue streams.
As the world moves away from fossil fuels, we don’t want utilities companies to emerge as the new dinosaur.

5. Are organisations ready in terms of their digitalisation strategies?

Some more than others for sure, but I think we’re all learning as we go while we chart this new course. But one thing is clear: to support the outcomes we want, we need to start with a solid architecture.

AI as you likely know, is a data hog, so at the core of IFS’s AI strategy is having a solid data foundation. On top of this foundation, the technical components are built that broker the queries and transactions of the themes and use cases specific to utilities. 

And the whole point of AI…at least in my opinion…is that it enables us mere mortals to engage with complex technology, and large amounts of data, in a natural and user-friendly way. So, at the interface level there are integrations with Microsoft’s Co-Pilot as well as native IFS interactions supported.
Now as we explored earlier, there are dozens of use cases, with more being developed all the time for this technology! But to showcase just a couple in particular, here are 2 that I think are really exciting and are delivering some truly exceptional results.

First, Asset Predictive Maintenance or APM.

Now you can run this with historical data alone. But when you combine APM with IoT and AI you get real-time performance monitoring capabilities, as well as insights and recommendations on corrective actions to be taken.
Adding on Schedule Optimization and Field Service…you have a truly efficient, end to end business process, supported by cutting edge technology, around your asset lifecycle management strategy.

And when you’re dealing with millions of assets in the field…and data coming in from multiple sources…AI driven Anomaly Detection is a real game changer. Saving not only time and money…but improving safety, reliability, and customer satisfaction metrics.

Next up, there’s AI backed Schedule Optimization. IFS’s award winning PSO solution can deliver exceptional results like a 35% reduction in travel time, or a 300:1 ratio of Dispatchers to Technicians, or an almost 50% reduction in contractor spend.

Now we can argue about whether a utility is ready to embrace AI and Automation to those levels or not…but the technology and capability is there, ready when the industry is.

6. What does the future look like, 10/20 years from now, in terms of the roles of humans and machines. What does the optimal relationship look like?

I don’t think technology will ever replace people, it will just make us smarter, more efficient and ideally more satisfied with our work life.

But change is hard and we need to manage and help people through this change.

And that will likely include re-training and career path planning, so no-one feels threatened by, but instead embraces technology as an enabler– taking us into our future.

Carol Johnston, VP, Energy, Utilities & Resources, IFS

Carol Johnston has over 20 years of product marketing and product management experience delivering mobile workforce management, outage management and meter reading solutions for utilities. She is a member of the advisory/planning committees for DistribuTECH International and Western Energy Institute and has served as track chair for SAP for Utilities.

Stay in the loop with Future of Utilities

Subscribe to our newsletter to receive news, insights and special offers.