AI & machine learning to play an increasingly important role in reaching net zero

Technologies, such as AI (Artificial Intelligence) and Machine Learning, are increasingly touted as essential tools for businesses and individuals to reach net zero
AI & machine learning to play an increasingly important role in reaching net zero
Like

Source: Pixabay

AI playing an increasing role in sustainability as the world digitalises

With the world facing increasingly short timescales to meet the goals set out by the Paris agreement, as well as the numerous national-level net zero goals set for 2030 rapidly approaching, technologies such as AI (Artificial Intelligence) are seeing heightened interest as a means to reduce emissions and improve efficiencies. By using AI alongside an increasingly digitalised world, the hope is that efficiency savings and optimisations will help to lower emissions overall. The complex monitoring and calculations carried out by AI can help establish total emissions across broad areas such as supply chains, business operations, and product lifecycles.

A recent study by IBM has highlighted just how important AI is becoming to those in the supply-chain segment, particularly relating to the challenge of sustainability. Undertaking a survey of 1,500 CSCOs and Chief Operating Officers (COOs), the study found that these individuals and their organisations are increasing investments in automation, AI and intelligent workflows, ecosystems and sustainability, as well as reimagining their supply chain operations.

Crucially, almost half (47%) of surveyed CSCOs said they have introduced new automation technologies in the last two years, and they ranked sustainability as their third biggest challenge in the next few years, trailing only supply chain disruptions and technology infrastructure.[i]

IBM also identified a sub-group within their study, the so-called “Innovators”, who represent roughly 20% of respondents and are those who stand apart for accelerating their data-led innovation. It found that “Innovators” are modernising their technology infrastructure – 56% of those respondents are currently operating on hybrid cloud, and 60% are investing in digital infrastructure to scale and deliver value. This group is already outperforming peers on key metrics, including reporting 11% higher annual revenue growth.

In a press release, Jonathan Wright, IBM Consulting Global Managing Partner, Sustainability Services and Global Business Transformation, has said: “To effectively combat the unprecedented supply chain stressors like inflation, it’s imperative that CSCOs focus on using analytics, AI and automation initiatives to build intelligent, resilient, and sustainable supply chains,” said. “Automation and AI can enable CSCOs and their organisations to collect data, identify risk, validate documentation, and provide audit trails, even in high inflationary periods, while also managing their carbon, waste, energy and water consumption.”[ii]

Expansion of AI and Machine learning comes with challenges

With mass digitalisation comes fears around energy usage, data storage, and waste. Last year IDC forecaste that the number of connected devices will reach 55.7 billion by 2025, of which 75% will be connected to an IoT platform[iii]. Whilst deep learning can be harnessed to help solve sustainability challenges, there is also the risk that vast expansion of AI and Machine Learning capabilities could see a company’s carbon footprint swell in size, for example, when considering the energy usage requirements.

As Venturebeat note- “Deep learning algorithms require a colossal amount of power when they analyse data. If left untouched, this could be a vicious cycle where, simultaneously, AI techniques are being used to identify potential environmental hotspots while the machines themselves consume huge amounts of power — thereby offsetting the positive impact.”[iv]

Yet, as the article argues, without help from technology, outlining sustainability goals would be a limiting and difficult exercise. Today’s businesses struggle with quantifying the risk of climate change, especially when it comes to digital transformation. “In fact, only 43% of global executives say they are aware of their organisation’s IT footprint. Data analytics and AI offer a solution to this challenge, as they provide meaningful insights across industries to understand where those gaps exist and thus can help companies incorporate more sustainable practices.”[v]

As business digitalises, there are those seeking a ‘cleaner cloud’

With our ever-digitalising world, the uptake of cloud computing is ever-growing. As we discussed in our blog earlier this year, cloud computing is an increasingly depended upon and fundamental component within the process of delivering computing services via the internet. It is therefore unsurprising that the requirement for electricity and computing equipment to facilitate demand is also on the rise. It is estimated that in 2020 data centres accounted for 1% of global energy usage,[vi] lessened in part by rapid improvements in energy efficiency, which have helped to limit growth in energy demand. Further, data centres themselves create vast amounts of heat, meaning electricity-hungry cooling systems must also be in place. Concerningly, Greenpeace finds from its estimates that “by 2025, the technology sector could consume 20% of the world’s total electricity; this increase from 7% currently (2019) is attributed to the expansion of cloud computing and the further development of new technologies, such as artificial intelligence, which require a great deal of computing power.”[vii]

The matter of adopting a ‘cleaner cloud’ is something which UK firm Kaluza has considered. In August this year, the company announced its use of real-time emissions data from Google Cloud, helping to inform its own internal carbon footprint tools. Essentially, Kaluza’s tool enables it to study Google Cloud data and track the impact of its cloud usage; this then enables it to identify areas where usage can be drastically reduced. Not only has Kaluza used the technology to create greener software through the introduction of a Green Development handbook, they have also been able to consolidate a number of large BigQuery queries into a single query at a greener time of day and location. This has resulted in a 97% reduction of emissions. Essentially reducing the amount of CO2 from 200kg to 6kg every time the company runs that query.

Kaluza’s sustainability manager Tom Mallett said in a blog post at the time: “Choosing a cleaner cloud and a cleaner cloud region to run workloads is one of the simplest and most effective ways we can reduce our carbon emissions. Fortunately, Google Cloud publishes carbon data for all cloud regions. This includes the average percentage of carbon-free energy consumed in that particular location on an hourly basis and the grid carbon intensity of the local electricity grid. By digging into the data, we can identify cloud waste and take action. For example, while many of our workloads have to run throughout the day, not all of them have to run at certain times. This creates potential for optimisation. We’re using data from Google Cloud to understand the state of our workloads. By combining this information with carbon intensity data from the grid, we can identify and reschedule workloads to lower intensity times, and have a positive impact on Kaluza’s emissions.”[viii]

References

[i] IBM Study: Supply Chain Leaders Are Investing in AI and Automation to Navigate Supply Chain Uncertainties and Improve Sustainability - Australian Associated Press (aap.com.au)

[ii] Ibid

[iii] Future of Industry Ecosystems: Shared Insights & Data | IDC Blog

[iv] Ibid

[v] Ibid

[vi] Data Centres and Data Transmission Networks – Analysis - IEA

[vii] Uncovering the Environmental Impact of Cloud Computing | Earth.Org - Past | Present | Future

[viii] Kaluza uses Google Cloud to help people reduce emissions | Google Cloud Blog

Please sign in or register for FREE

If you are a registered user on Zero Carbon Academy, please sign in