Representative End User Clients
Representative Automation Clients
Representative Software Clients
ARC’s European Industry Forum, part of the successful series of worldwide conferences in Europe, America and Asia, has been held in Sitges (Barcelona), Spain on May 6-8, 2024. The event offered exclusive presentations and workshops on strategies and case studies in line with this year’s topic “Managing Digital Transformation in the Age of AI, Open Architectures, and Sustainability” to its 150 international participants from over 20 countries.
AI in Automation
The focus of the session on “AI in Automation” was on keeping a realistic view on industrial grade AI technologies. It provided valuable insights into the current state and future potential of AI in automation, highlighting the need for collaboration and a common understanding of industrial grade AI.
George Hancock from OctaiPipe presented his company’s Federated Learning platform, based on a distributed learning technique that allows multiple devices to collaborate on a shared model without exchanging their data with other parties. This approach preserves the privacy and compliance of the data owners and reduces the communication and computation costs compared to centralized learning. However, it also faces challenges such as heterogeneity, complexity, labelling, and connectivity.
Nicolas Spiegl from UReason gave an insightful presentation about the "Pragmatic use of AI and LLMs". He argued that most of the data produced by production facilities, plants, and critical infrastructure are not yet used to optimize the processes. UReason uses a network of different algorithms to convert data inputs into valuable results, as well as large language models (LLMs) to enhance the functionality of their tools. However, he admitted that customers do not really care about the fancy AI models behind the solution, as long as they work well and provide useful insights. He concluded that a pragmatic and easy start is by far better than not starting at all due to complexity.
Sathyamurthy Sundaram from Voith Group illustrated in his the challenge of user acceptance where AI technologies are concerned. He suggested that as a company, we should think big, but start small, when introducing AI solutions to customers, whether internally or externally. He advised to gradually increase the level of autonomy and let the people involved get used to it, rather than imposing it on them. Sathya also explained how AI technologies can help companies to analyze the data from production processes and identify the root causes of problems or inefficiencies, as well as prevent them from happening in the first place.
Boris Scharinger from Siemens presented the idea of industrial grade AI foundation models, which could be game changers for the industrial AI market. The models could learn from various industrial data sources and domains and adapt to different industrial scenarios. The key to creating them would be to establish data partnerships and pools for specific use cases, where the data could be shared and used securely and ethically, while respecting the data providers. He concluded that companies collaborating on data sharing will have an edge in the industrial AI market, as they will benefit from the power and efficiency of the industrial foundation models. Boris is convinced that industrial AI would eventually become more scalable and profitable once the technology improved its reliability and robustness.
ARC’s European Industry Forum
We like to thank our sponsors, who supported the ARC EIF 2024, and all our speakers and presenters, who made the event interesting, exciting and memorable. The next Forum will take place in Sitges (Barcelona), Spain, on May 5 – 7.
For more information, please contact Ann-Kathrin Blech (mailto:[email protected]).