The UNS acts as the single source of truth, integrating and contextualizing data from across the manufacturing landscape. Functioning as a hub-and-spoke architecture, it breaks down silos by enabling real-time data flow between all connected systems, whether horizontally across systems at the same level or vertically across hierarchical levels. This unified approach organizes and contextualizes operational data into a standardized structure, creating a reliable foundation that advanced technologies — such as AI, advanced analytics platforms, digital twins, industrial IoT (IIoT) systems and predictive maintenance tools — can leverage to generate actionable insights.
Boosting operations with AI
Operational staff are the backbone of any manufacturing facility. They know the ins and outs of the day-to-day operation of the equipment and processes but often lack the tools to interpret vast amounts of data at their disposal.
That’s essentially what can happen when LLMs are integrated with data contextualized in a UNS. This combination empowers subject matter experts (operators, technicians, supervisors, engineers and lab analysts) with intuitive tools to navigate complex data, enabling them to uncover actionable insights faster and make more informed decisions in real time.
Imagine an operations supervisor asking an industrial version of Gary:
- Why did we have unexpected downtime during the night shift?
- What adjustments can improve yield in the next production run?
- How can we reduce energy consumption during peak hours?
Instead of spending hours sifting through endless logs or navigating complicated dashboards, the AI processes the question, retrieves relevant data via the UNS and delivers clear, actionable answers. It’s like having Gary on the plant floor — always ready to assist and always equipped with the right information.
Even with this advance, it’s still the expertise of the plant floor staff that steers Gary’s responses, ensuring the insights are meaningful and aligned with real-world operations.
Simplifying on-prem AI infrastructure
As exciting as cloud-based AI services are, they’re not always the best fit for manufacturing environments. Concerns about data security, latency and compliance are driving a shift toward on-premises AI as a more practical solution. Manufacturers require infrastructure capable of supporting the demanding workloads of advanced models like LLMs.
Cisco, a pioneer in AI infrastructure, has developed validated designs that integrate compute, storage and networking into modular configurations. These systems, known as AI Pods, offer a full-stack solution that allows manufacturers to deploy advanced AI technologies without the complexity of building custom systems from scratch. With their modular design, AI Pods enable organizations to start small, evaluate their needs and seamlessly scale as AI adoption grows.
By combining industrial ETL, unified namespace, and on-prem AI infrastructure, manufacturers can create a seamless, secure and efficient environment for AI adoption. It’s all about getting these advanced tools into the hands of the people who can take action and make a real impact.
So, why not bring a little bit of Gary to your plant floor? You might be surprised at how much you and your people can accomplish.
Dan Malyszko is vice president at Malisko Engineering, a certified member of the Control System Integrators Association (CSIA). See Malisko Engineering’s profile on the CSIA Industrial Automation Exchange.
More AI co-pilot and unified namespace coverage from Automation World:
