Article
    by Jukka Kostiainen, Value Architect at Nortal

    Real-time data + real AI = real solutions to factory floor problems

    Bust the hype: GenAI is not a hammer for all nails.

    Service

    Industry Data and AI Manufacturing

    Industry

    Data and AI

    Every Factory Lead and Head of Production understands the frustration of receiving yesterday’s production report, only to discover problems that are already too late to address. When you oversee industrial operations, the pressure is constant: profit margins are narrow, equipment must operate continuously, and safety cannot be compromised. Any delay in data can result in, e.g., lost output, wasted energy, and rising costs, ultimately eroding your business while reducing profitability.

    Generative AI (GenAI) and large language models (LLMs) are today’s hammer that makes every problem look like a nail. While GenAI can write text, create images, and even answer questions like a human, it is not a complete solution for the challenges faced on your factory floor. Your ultimate game-changer lies elsewhere: in real-time data combined with AI-driven analytics, including Machine Learning (ML), and – yes – in some cases, GenAI as well.

    Technologies that enable the effective use of real-time data have been available for quite a while. However, in our experience, many Finnish factories, for example, still rely on batch data. This approach means that analyses and decisions are based on what happened yesterday, rather than on what’s happening now – let alone what’s going to happen later today. Too often, our customers try to navigate their operations by looking only at yesterday’s trends.

    What’s hitting your bottom line aren’t IT issues, but business bottlenecks

    Siloed, inaccessible data is often viewed as an IT problem. We have seen that this perception frequently stems from the fact that no single business area considers the challenges posed by data delays significant enough to warrant action. Despite this, the demand for real-time data is growing, particularly for production-related use cases that can drive margins. This trend is due to changes in the marketplace and the rise of low-threshold AI technologies, including, but not limited to, LLMs.

    For example, in fall 2025, Nordic electricity markets moved to 15-minute settlement periods, resulting in 96 price intervals per day instead of the previous 24. This change means that electricity prices now fluctuate four times an hour. This shift aims to improve flexibility and better integrate renewable energy sources. However, it also means that what was previously considered “good enough” planning is no longer adequate.

    Nortal recently worked with our customer, a large forest industry company, on a project to accurately forecast energy usage in the current, highly volatile 15-minute market. 

    in the current, highly volatile 15-minute market. This kind of real-time data and reliable forecasts of factories’ energy consumption would allow any industrial player to identify energy-intensive parts of its processes, optimize them, and balance demand, supply, and prices. Using outdated, inaccurate consumption forecasting, on the other hand, means industrial companies will either pay a premium for last-minute energy or lose money by selling their excess power at a discount. Purchasing energy, a day in advance is the most cost-effective option. Companies that effectively utilize real-time data and machine learning (ML) algorithms to forecast their 24-hour energy consumption can save significantly. With these capabilities, industrial players can optimize both their own energy usage and the energy they purchase.

    Another silent margin killer in industrial settings is unplanned downtime. A single broken machine can paralyze an entire production line. For example. at one of Nortal’s manufacturing customers, a failure in a processing line could result in thousands of kilos of food spoiling before anyone even has a chance to react. This not only results in wasted products but also incurs lost revenue and dissatisfied customers, which can have a ripple effect across the supply chain. Companies could save millions by gaining real-time visibility into what is happening at their manufacturing sites. Even better, they could utilize that data and AI to predict equipment health before a failure occurs.

    Real-time analytics can ensure that you have the right people in the right places at the right time, while also prioritizing their health. Picture this: a packaging line unexpectedly slows down. Do you have the flexibility to redeploy staff from another area? With real-time visibility into workforce allocation, you can achieve significant savings by facilitating more effective employee rotation.

    From a workplace health perspective, real-time data keeps you informed about such issues as unauthorized activity in restricted areas, compliance with safety gear rules in your factories, and potential collision risks when people and machines, like autonomous forklifts, operate on the same premises.

    The challenges described above aren’t just IT headaches; they represent business bottlenecks that can impact margins, efficiency, and competitiveness. The longer these issues go unaddressed, the greater the cost to your organization.

    Discover how we can help you leverage Microsoft Fabric for real-time, compliant data

    From hindsight and scribbles to foresight and confidence

    Instead of focusing on today’s challenges, imagine having a clear, real-time view of your factory. You know what’s happening right now (not just yesterday), and you could predict what will happen in the next hour or even tomorrow. This capability allows you to adjust production before problems escalate, optimize energy costs by forecasting consumption, proactively add staff at critical production lines or cells, and prevent costly breakdowns or accidents on the shop floor by identifying anomalies early.

    Here lies the magic of real-time data. It serves as the foundation for more intelligent technology use cases. AI, whether through machine learning or advanced models like generative AI, transforms that data into actionable insights and foresight.

    So, how does GenAI fit in? It is indeed powerful, yes, but it’s not a silver bullet. It can serve as an intuitive user interface for various tools, and one of its greatest promises in manufacturing lies in its ability to integrate structured and unstructured data. GenAI can transform messy, handwritten maintenance logs from past decades into clear, actionable instructions. This would reduce your factories’ reliance on experienced workers who may be the only ones to remember how to resolve disruptions on the manufacturing line or who have jotted down notes in a pile of notebooks.

    Once you have historical data organized effectively, and make real-time data, structured machinery data, and analysis tools available, you will create a dynamic digital assistant for your maintenance workers. The tool adapts to on-site events, enhances maintenance and troubleshooting efforts, and significantly increases the onboarding efficiency for new employees.

     

    At Nortal, we understand manufacturing and data. We’ve developed real-time data and AI solutions for companies like Outokumpu, a large forest industry company and Valmet. One key lesson we have learned is that investments in data often stall while waiting for that one “killer use case” that promises to transform everything.

    GenAI could be that game changer for your business, or it might be other forms of AI analytics like Machine Learning. However, we challenge you to consider a different approach: what if you invest in an AI-ready data foundation today rather than waiting for the ideal “killer application” to emerge? By establishing scalable, real-time data capabilities now, you can accelerate innovation, run rapid experiments, and harness the value of novel AI solutions ahead of your competitors.

    Explore our white paper and find out how you can be sure you’re taking advantage of the full potential of your industrial data.

    Why cyber exercising matters

    • Reveals critical gaps in technical controls, escalation paths, and decision-making workflows.
    • Fosters organisation-wide collaboration, improving coordaination and communication across all roles, functions, and levels. Builds confidence under pressure, giving participants, groups, and organisations muscle memory they can rely on.
    • Exposes participants to real-world attack techniques, improving detection, containment, and familiarity.
    • Strengthens regulatory and stakeholder alignment by stress-testing notification and reporting procedures in a simulated environment.
    • Fosters a culture of continuous improvement by turning lessons from exercises into actionable changes across people, processes, and technologies. 

    Talk to our industry experts

    Tell us how we can help. Our experts will be in touch.