How AI will Help Close the Sustainability Gap

How AI will Help Close the Sustainability Gap

Jim Chappell, Global Head of AI and Advanced Analytics, AVEVA, explains how different AI technologies will increasingly support sustainability transition across industrial world

The pursuit of profit is no longer at odds with environmental sustainability. Artificial intelligence (AI) is fast bridging the gap between the two sides, accelerating the scale and pace of sustainability solutions needed to address the worsening climate crisis.

AI increasingly plays a vital role in the world’s transition to greener operations.

Indeed, some 87% of industrial leaders see AI as a useful tool in the fight against climate change, according to a recent BCG survey. Approximately 43% envision leveraging the science in their own climate change efforts.

AI technologies are already supporting companies on the journey to achieving their net-zero targets. AI is already helping decarbonize hard-to-abate sectors.

While generative AI has dominated headlines recently, industries are also onboarding other types of AI technologies in different ways. They are using it to integrate new sources of renewable energy into production lines, drive greater productivity and efficiency, tap new insights for better decisions, and build more agile, resilient operations.

Which AI technologies are being used in industry today?

Let’s look at just four major AI technologies being used across the industrial spectrum today.

·        AI-powered predictive analytics can help businesses to anticipate demand, optimize supply chains, forecast asset anomalies, and streamline inventory levels in real time. Using statistical algorithms and machine-learning technologies, current and historical data can be analyzed to predict future events—including forecasting overall GHG emissions. As a result, costs and resource use are reduced, in turn lessening the environmental impact of overproduction and unnecessary resource consumption.

·        The next step is predictive asset optimization. Here, dynamic simulation tools, together with predictive analytics and advanced visualization, create a hybrid digital twin. Users gain a true 360-degree view of operational risks and can identify and fix problems earlier, as well as forecasting remaining useful asset life for maximized uptime, availability and profitability. When incorporated into the design of future assets, these insights activate a cycle of continuous improvement. In real terms, predictive asset optimization accurately forecasts performance degradation and greenhouse gas emissions in depth at a granular level.

·        Generative AI is perhaps the best-known way people encounter AI, both in their daily lives and in industrial applications. The technology has been around for more than half a century, but has now come into its own recently as massive large language models (LLMs) become available to the public. They enable operators to make sense of large knowledge sets quickly, or serve as a creative partner to support innovation—such as by mocking up asset design options according to specific parameters or creating engaging technical learning material. When used in conjunction with real-time data leveraging specializing software, it can also provide deeper insights into data, including assisting with the complex analysis of sustainability issues.

·        Grey-box modeling, among most cutting-edge industrial AI technologies, is just hitting the market. This combination of first principles simulation and AI (“white box” and “black box” models respectively) delivers the best of both worlds: modelling assets and processes in near-real time in order to improve system design and also get the most out of it operationally. One way it works is by allowing AI models to be integrated with traditional physics-based simulation via a drag-and-drop user interface. AI often runs faster than physics-based models and requires less tuning to set up. As a result, companies can get models up and running quickly, while using less CPU capacity and therefore requiring a smaller carbon footprint.

What gains is AI delivering for industries?

Industrial AI solutions contextualize key performance and sustainability data with artificial intelligence and human insight. Consequently, enterprises can unlock value and sustainability gains in many ways – right now and in the future.

In the energy sector, AI technologies are supporting the transition to renewables. Italian multinational Enel has committed to decarbonizing its energy mix by 2040. It has already installed more than 50 GW of installed renewable capacity. To speed the transition, Enel has deployed an AI-infused asset performance management software teamed with predictive analytics. Data silos have been eliminated, accelerating decision-making and delivering efficiencies across the business ecosystem. The energy leader can now predict asset failures and ensure steady power supplies, and is en route to achieving a fully autonomous plant.

Likewise, another global energy company is using predictive asset optimization to improve reliability and cut maintenance costs. Just one catch helped detect a performance anomaly in heat recovery pipes at a cogeneration unit five months in advance, saving significant costs. Since 2019, the software has detected more than 1,700 asset performance anomalies. More than $37 million has been saved, unplanned downtime slashed and resource use reduced, improving environmental impact.

AI is delivering similar gains in other hard-to-abate sectors such as cement, which generates 6% of all manmade emissions.

Oyak Cement, whose operations extend from Turkey to Portugal, Cape Breton, and West Africa, uses an AI-infused edge-to-cloud data management system to replace 30% of its fossil fuel-sourced energy with renewable sources, as well as to cut energy use. For every 1% reduction in energy used, the company saves €5-7 million. With real-time information to hand, it also reduces CO2 emissions, ensuring regulatory compliance.

At a more fundamental level, AI is helping boost resilience against the worst effects of climate change. With rising global temperatures, the city of Salem in the Pacific Northwest state of Oregon has witnessed more outbreaks of toxic algal blooms on catchment lakes and rivers.

With the help of a multi-tenant, cloud-native data management platform, city authorities have aggregated various sources of data—from algae levels to water depth and weather and satellite data—into one central hub. Intelligent predictive insights now alert them to increased algae and cyanotoxin activity two weeks before they occur, enabling teams to preserve water quality, safeguard ecosystems and ensure that Salem’s five million residents have safe drinking water.

Industrial AI is essential to the sustainability transition

The industrial world is now entering what has been called an AI revolution.

To view AI merely as a tool for boosting profits is to severely underestimate—and underutilize—the capabilities of this potent science. As industrial enterprises work to embed sustainability at the core of their operations, AI can empower them to align business success with environmental responsibility. 

AI won’t solve the climate crisis, but it can unlock greater value for industries while improving sustainability – although success in each case requires both time and effort and depends on how the technologies are applied.

Overall, however, the marriage of sustainability and profitability is no longer an elusive goal. With its capabilities to optimize processes, enhance efficiency, and foster eco-friendly practices, AI serves as a binding element that brings together these seemingly separate goals to create the future we need.

Sustainability Middle East News