The democratisation of data is a remarkable opportunity for the industry

In an exclusive interview, Rob McGreevy, Chief Product Officer of AVEVA, sat down with Energy Connects to talk about how opening up application development and enabling a community of data experts can create new business opportunities, the contribution of small language models in the development of AI for energy, and the way forward for making highly complex energy value chains more productive and sustainable at the same time.

• What are the opportunities arising out of the true democratisation of technology?

The democratisation of data, applications, people, and ideas has been, it’s actually a remarkable opportunity. I think just democratising data enables innovation for people to find new and innovative solutions in areas like sustainability, energy management, and productivity improvements. So one very simple thing is just by making data available to people who have curiosity, want to engineer, want to explore, want to invent, it becomes more feasible.

I think the idea of democratising application development is called citizen developers – non-programmers who can write and develop things. I think putting the technology and data in the hands of people that just want to solve problems but maybe aren’t as technical or as computer science focused or engineering focused – I think is another great opportunity. I think the evidence of some of this is in the success stories and case studies we’ve showcased at AVEVA World in October last year – where new innovative solutions like Brilliant Planet comes to mind. These folks have a tremendous amount of data about an organic process of growing algae. And yet, they found a way to use science to evolve and manufacture this algae at a very effective rate. They use technology to share data, they have data scientists that also look at the data and innovate on it to find better ways of increasing yield, for example.

The further sharing of that data with regulators is necessary so you can capture the carbon and verify that the carbon credits can be essentially achieved. So when you really think about any one of these use cases, they start with democratising data, enabling the community of experts around there, and also create new business models.

• You and [AVEVA CEO] Caspar Herzberg have both spoken extensively about not just big data but also the importance of small data and how it adds up. Can you talk to us about what is the importance of all segments of data in the context of the predictive models that you are building?

I think a recurring theme that comes up is that not only is there a lot of data collected and captured, but there’s also this paradox that there’s also a lot of equipment, machinery, and processes that are not instrumented that we don’t actually have data captured and collected to. I think there’s a huge amount of opportunity to further develop all the information that’s out there. But at the same time, it’s important that this data becomes somewhat specialised, or smaller in a sense.

How do you extract the meaningful bits out of this information, so that it becomes practical and useful? So as an example, you have lots of flow rates or vibration or temperature data. How do you turn that into a meaningful set of data that says, yes this is good news that’s vibrating at this level, or it’s bad news or this temperature profile yields a better sustainable solution versus a lower temperature.

It’s interesting you bring up the point about the small and big data in the realm of AI analytics and language models. As an example, there’s a lot of emphasis around not only the large language models, but the small language models. Large language models are obviously quite enormous because they consume and they train on huge volumes of data. Think of Wikipedia, encyclopedias, as well as subjects like science.

Small language models on the other hand take the same techniques, but they’re very specialised. So you create a small language model for maybe process optimisation. That language model may not be able to tell you the history of England, but it could tell you about rotating equipment or analysis. I think we’ll continue to see highly specialised implementations come out of the world.

And the form of large, small or specialised language models, and similarly the data –whether it’s small data, large data, big data, whatever you want to call it – will also be curated specially based on these different use cases. So it’ll be a very interesting world, but I think for industrial, for manufacturing, what we do in sustainability, it’ll give us great results that are highly tuned to the markets that we serve.

• How is AVEVA Connect harnessing data and AI as a platform to help make the energy industry more sustainable?

Our belief is that in the energy sectors in particular there’s highly complex value chains but in order to keep producing and providing energy to the planet and to the world at large, we need to continue to expand and grow that.

At the same time, we also have to improve the sustainability of it … We’re saying we need more energy, and at the same time we’re saying we must be more sustainable. And we do have to do both. And AVEVA Connect certainly helps to take advantage of that: at a very basic level, capturing and collecting information around water, air, steam, gas, and electricity usage – the basic wages as we like to say.

These are all elements that have sustainability with them. How do I make more intelligent use of the water? How do I use less electricity? How do I conserve gas consumption? Connect becomes this platform for capturing and collecting all of the elements that are necessary to drive sustainability goals.

On the other hand, it also helps optimise the processes to produce less emissions and to operate with a lower carbon footprint. And some of the tools that Connect enables are not only the data capture and collection, but the AI we talked about. We can use predictive models to say, look, if you run this particular piece of equipment at this rated speed, you can expect this amount of CO2 emissions. And if you make these changes, perhaps you optimise the production to solve lower CO2 emissions, that can give you flexibility and achieving sustainability goals.

So I think as an industry, Connect can help close a lot of the gaps and enable the energy markets at large to move forward in a safe and sustainable way. When it comes to the energy industry and obviously almost all industries, the energy transition and people talk about the technologies that are not yet here that is needed perhaps sooner in the future, Connect can help scale up and achieve net zero goals.