Starting Points for Leveraging Industrial AI
Industrial AI is already helping companies accelerate their digital transformation across a range of sectors, from advanced manufacturing to infrastructure and energy systems.
And the future potential is enormous. But to realise the benefits, there are two things we have to take into consideration.
Ensuring industrial grade and making sense of the hype
Firstly, we must ensure the technology is industrial grade, designed to optimise performance and ensure safety and reliability in highly constrained industrial settings.
This is different from the Generative AI that every business is adopting, which empowers creation through natural-language prompts.
In summary, industrial grade AI must be robust, democratised and deliver applications with real purpose in industry.
Robustness is essential as industrial processes must be reliable and ensure the highest levels of security and trust, or adoption will never scale.
Democratisation means we can empower non-AI experts to access and operate the technology.
And through purpose, the deployment of AI in industry has to support the goals of the company, from increasing productivity to sustainability, while being simple to scale across different processes at speed.
While generative AI can already achieve many things with competence, the answers an LLM gives are not always predictable and can give different answers from the same prompt.
These tools are great at understanding how to answer open queries, but they struggle to consistently answer right-or-wrong maths problems involving complex calculations. This means that any AI used in industry must be industrial grade.
At Siemens, we’ve worked in AI for decades and this is why we’re developing the first Industrial Foundation Model, which can understand the language of engineering and ensure the benefits of Generative AI can be applied in the industrial world.
Secondly, we must make sense of the noise and hype around AI.
Much of it is warranted. But it’s created a wariness of the technology and indecision on how best to get started, not least because of the number of different tools and platforms that have emerged over the last 12 months.
So where to begin?
In the conversations I have, it’s clear that businesses often feel overwhelmed by the options available to them and are stuck on where to begin.
Outlining the practicalities is always helpful, so here are five starting points to help companies leverage industrial AI to deliver measurable impact:
- Ensure you have a purpose
An important first step is ensuring that you can identify a purpose before jumping into the complexity of integration, and that means investing time into considering the right use cases and applications.
Our work with Yorkshire Water is a powerful example. Working alongside the University of Sheffield, the focus was on safeguarding natural water systems by preventing pollution risk.
The project concentrated on integrating AI into the monitoring of 55,000 km of sewers to increase the speed and accuracy of spotting overspills, and to learn the unique rainfall runoff pattern for each combined sewer outlet site so Yorkshire Water could predict blockages up to two weeks ahead.
This replaced legacy methods that were prone to false alarms and late detections, cutting pollution incidents by half and realtime operational improvements that weren’t achievable before.
- Start with small, strategic experiments but design to scale from the outset
The next step should take that purpose into creating a culture of strategic experimentation in engineering teams.
This could start with a small, scalable pilot that should not only focus on providing a strong case for further investment and scaling digitalisation, but in building skills and confidence in using the technology.
Take Entocycle, a pioneering manufacturer of insect farming technology focused on creating a more circular food system that will reduce the UK’s food waste and reliance on imported soya.
We used AI to model a small but groundbreaking hi-tech insect farming pilot in the arches under London Bridge and optimise how the space was used.
The technology was also integrated to enhance the yield by continually monitoring the health and size of the colony, ensuring the insects get the right amount of food and that precise temperature and humidity levels are maintained.
- Build the right partnerships
We believe that the complexity of the world’s challenges can’t be solved by the capabilities of one organisation on its own, which is why we increasingly work with partners to deliver outcomes for customers.
Collaboration is critical and industrial AI succeeds when it’s built on partnerships that combine deep domain expertise with proven digital platforms.
At Siemens, we work with a range of partners, from Microsoft and AWS to pioneering AI start-ups and scale-ups, to translate generative AI into industrial-grade tools that can be applied to complex processes and existing hardware.
For any business, collaboration with technology providers who understand your sector’s regulatory and operational context is essential.
- Establish trust in safety-critical applications
Robust governance and security is vital, particularly if you’re deploying AI in environments where safety, continuity and compliance are non-negotiable.
This should include comprehensive data protection, explainable models, clear validation frameworks, and importantly, human oversight for decision making and verifying actions across a process.
Building in structured human intervention is key to scaling its use responsibly and in establishing trust among employees, customers and other stakeholders.
And this means that ensuring you have the right skills in place to support AI integration will be key.
-
Recognise that now is the time for action
The UK‘s track record for adopting new technologies has been mixed and while many businesses have embraced digital tools, overall investment in digitalisation has lagged behind international peers.
Failure to take advantage of AI now will risk that gap widening further, which will only serve to keep the brakes on productivity.
At an organisational level, now is the time to harness the technology or lose ground on global competitors.
By defining a clear purpose, forging the right partnerships, experimenting strategically and embedding trust, businesses can unlock the benefits of AI safely and effectively.
At Siemens, we combine the real and digital worlds to empower our customers to accelerate their digital and sustainability transformation, helping them uncover new possibilities and create more resilient, efficient businesses.
In a world of constant change, taking the first step is the most important. Now is the time to act.
Sprint Campaign: Industrial AI
From predictive maintenance to advanced process automation and smarter supply chain management, Industrial AI is the application of AI technologies in industrial settings to transform operations across various sectors—including manufacturing, and robotics. This campaign will showcase the game-changing potential of Industrial AI, and how it can solve industry challenges, drive efficiency, increase productivity, boost innovation, and redefine the future of industrial operations.
Event Round-ups
Authors
Dan Smalley
Head of Industrial AI – Digital Industries , Siemens UK & Ireland
