As a result, it is not designed to support unified, contextualized, and scalable data management across an organization.

And that’s why Industrial Data Platforms are essential for scaling digital transformation. 

To help organizations understand what makes such a platform effective, David Ariens and The IT/OT Insider team created the Industrial Data Platform Capability Map, outlining the seven key capabilities every platform should have:

1. 𝐂𝐨𝐧𝐧𝐞𝐜𝐭𝐢𝐯𝐢𝐭𝐲 – a secure and scalable connectivity layer to integrate different data sources into the Industrial Data Platform.

2. 𝐂𝐨𝐧𝐭𝐞𝐱𝐭𝐮𝐚𝐥𝐢𝐳𝐚𝐭𝐢𝐨𝐧 & 𝐃𝐚𝐭𝐚 𝐌𝐚𝐧𝐚𝐠𝐞𝐦𝐞𝐧𝐭 – delivering data enriched with the right context, so users don’t have to gather and piece together information from multiple sources manually.

3. 𝐃𝐚𝐭𝐚 𝐐𝐮𝐚𝐥𝐢𝐭𝐲 – Detecting and fixing data issues in your pipeline, from sensor to final report.

4. 𝐃𝐚𝐭𝐚 𝐁𝐫𝐨𝐤𝐞𝐫 𝐚𝐧𝐝 𝐒𝐭𝐨𝐫𝐞 – The ability to ingest, store, and manage contextualized data at scale, enabling efficient data subscription and large-scale querying. 

5. 𝐄𝐝𝐠𝐞 𝐀𝐧𝐚𝐥𝐲𝐭𝐢𝐜𝐬 – The capability to perform analytics and machine learning within the data platform, or at the edge, close to where the data is generated.

6. 𝐕𝐢𝐬𝐮𝐚𝐥𝐢𝐳𝐚𝐭𝐢𝐨𝐧 – The capability to deliver high-quality, contextualized data to users through intuitive and accessible interfaces for fast, informed decision-making.

7. 𝐃𝐚𝐭𝐚 𝐒𝐡𝐚𝐫𝐢𝐧𝐠 – The capability to openly expose platform data to external users and applications through standard interfaces and integrations.

In the latest episode of the AI in Manufacturing podcast, I sat down with David, Founder of IT/OT Insider, to dive deeper into these capabilities and how organizations can implement them.

We also discussed the IT/OT Academy, an online training program designed to help IT and OT professionals build a shared vocabulary, framework, and collaboration strategy to move digital initiatives beyond pilot projects and into full-scale plant deployment.

ABOUT FOLIO:

Folio’s AI platform empowers industrial sales and application engineers by turning technical specs, configuration data, and application info into instant answers, recommendations, and agentic workflows, speeding work, cutting errors, and boosting revenue for industrial manufacturers and distributors. Learn more at www.folio.build

ABOUT FAY:

Fay Goldstein is the Co-Founder and CEO of Folio, an AI-powered platform that transforms how manufacturers and distributors sell and support complex and technical industrial product portfolios. Before founding Folio, she spent her summers managing direct and online sales at local automotive AC condenser and compressor shop, led strategic GTM and communications at an automotive telematics data company, and worked at an early-stage venture capital firm, where she supported dozens of early-stage startups on their initial GTM and communication strategies. Fay graduated magna cum laude from Florida International University and holds an MBA from Reichman University.

CONNECT WITH FAY 🌐 Website: https://www.folio.build/ 💼 LinkedIn: https://www.linkedin.com/in/faygoldstein/

Here’s a sneak peek at what we cover:

1. Predictive Maintenance for Machinery

• AI helps manufacturers predict equipment failures before they happen, reducing downtime and saving costs. With predictive maintenance, companies can transition from reactive to proactive maintenance, leading to longer machine life and fewer unexpected breakdowns.

2. Quality Control and Defect Detection

• AI-powered visual inspections now identify defects faster and more accurately than human inspectors, ensuring product consistency and quality. We discuss how AI-driven quality control is reducing waste and improving overall customer satisfaction.

3. Supply Chain Optimization

• AI tools are optimizing supply chains by predicting demand and adjusting inventory accordingly. In the episode, we break down how smarter supply chains are helping manufacturers avoid bottlenecks and reduce delays, especially during unpredictable market conditions.

4. Enhanced Worker Safety

• From monitoring working conditions to analyzing data for potential hazards, AI is making factory floors safer. Learn how wearable technology and smart sensors are helping manufacturers reduce workplace injuries and improve employee well-being.

5. Energy Efficiency and Sustainability

• AI is enabling manufacturers to cut down on energy usage and reduce their environmental footprint. This is a critical step as companies aim to meet sustainability goals and reduce costs.

🎧 Tune in to the full episode to discover how AI is reshaping the future of manufacturing—and what it means for businesses aiming to stay competitive.

Vast amounts of unorganized data collected from various sources often become "data swamps," making it difficult to extract meaningful insights and generate value.

This overwhelming complexity hinders decision-making and slows down innovation.

Additionally, the analytics tools currently available are often too complex and static for domain experts to use effectively, leaving them without the critical insights needed to improve processes, optimize production, and make informed decisions.

AI assistants offer a promising solution by bridging the gap between complex data sets and user-friendly interfaces.

They transform unstructured data into actionable insights accessible to everyone in the organization.

To learn more about the application of AI assistants for advanced manufacturing data analytics, I sat down with Stefan Suwelack, the CEO and Co-Founder of Renumics.

Episode Guest: Gilad Langer, Head of Digital Transformation Practice at Tulip Interfaces Host: Kudzai Manditereza Publication Date: [Insert Date]

Episode Summary

Manufacturing systems are complex adaptive systems that require a fundamentally different approach to AI implementation than traditional monolithic architectures. In this episode, Gilad Langer draws on 30 years of manufacturing experience—including PhD research that laid the groundwork for Industry 4.0—to introduce a composable agentic framework specifically designed for frontline operations. He explains why adaptability has become a competitive necessity in today's disrupted markets and how multi-agent systems can transform innate factory equipment into intelligent, communicating entities. The conversation covers practical implementation strategies, the artifact model for structuring manufacturing data, and why cultural change remains the biggest obstacle to agentic AI adoption.

Key Questions Answered in This Episode

• What is an agentic AI framework for manufacturing and why do factories need one?
• How do complex adaptive systems apply to manufacturing operations?
• What are the five pillars of composability in manufacturing?
• How should manufacturers structure their data for AI agents using the artifact model?
• What is the difference between staff agents, builder agents, and artifact agents?
• How do you implement agentic AI in a brownfield manufacturing facility?
• Why do traditional MES systems fail to deliver the adaptability modern manufacturing requires?

Arshan Poursohi brings a rich background in robotics and research, having worked with Sun Microsystems, Google Research, and Toyota Research. His experiences exposed him to global labor challenges, such as aging workforces and a shortage of young people entering physically demanding jobs. Recognizing the urgent need to address these issues, Arshan founded Third Wave Automation to apply modern AI and robotics solutions to material handling in warehouses and manufacturing environments.

The discussion highlights how AI and robotics are revolutionizing the manufacturing industry by automating "dull, dirty, and dangerous" tasks.

Key benefits include:

• Increased Productivity: Operators can manage an entire fleet of autonomous forklifts from a single workstation, significantly boosting the number of pallets moved per day.
• Enhanced Safety: By removing operators from hazardous environments, the risk of workplace accidents decreases.
• Immediate ROI: Third Wave Automation's as-a-service model allows companies to see immediate returns, with predictable uptime and reduced labor costs.

Third Wave Automation introduces the concept of shared autonomy or human-in-the-loop machine learning. Unlike fully autonomous systems that aim for "lights out" operations, this approach keeps humans in the loop:

• Collaborative Operations: Robots perform tasks but can recognize when human intervention is needed, such as when a payload is precariously positioned.
• Continuous Learning: Human inputs help train the AI models in real-time, improving performance and adapting to new scenarios.
• User-Friendly Interface: Operators control robots using familiar tools like steering wheels and screens, making the system accessible even to those without advanced technical skills.

Arshan shares a success story involving a customer who deals with large bags of powder prone to shifting—a challenge for automation due to safety concerns. Using shared autonomy, the autonomous forklifts learned to handle these unstable payloads safely and efficiently:

• Adaptive Learning: The AI models quickly adapted to recognize when human assistance was needed.
• Safety Assurance: Operators could intervene remotely, ensuring safe handling without halting operations.
• Productivity Gains: The customer saw immediate improvements in efficiency and safety, validating the effectiveness of Third Wave Automation's approach.

Listen to the full episode to learn more.

They assume stable product lines.

They assume you have time for lengthy implementations, tolerance for complexity, and operators who can navigate digital forms while running production.

But here's the challenge. Today's manufacturing reality is different:

⇨ Markets demand the flexibility to shift from 1.5-liter bottles to 1-liter bottles overnight

⇨ Low volume, high mix production is now the norm

⇨ Tribal knowledge is retiring faster than it's being captured

⇨ Workers stay 2-3 years, not 20, making traditional training models obsolete

The cost of this disconnect?

❌ Frontline workforce unable to contribute operational intelligence at scale

❌ ROI delayed by complexity, not capability

❌ Two-year deployment cycles for basic systems

❌ Digital initiatives stuck in pilot purgatory

That's why leading manufacturers are rethinking execution from the ground up, shifting from monolithic systems to AI-native, human-centric platforms built for today's workforce reality.

This new approach is effective because it’s built with an AI-native mindset, not a digitized version of paper-based processes

✅ AI-generated SOPs from video, cutting engineering time by 80%

✅ Learning systems that surface troubleshooting guidance from historical fault data

✅ Human-centric design that captures operational intelligence without disrupting workflows

✅ AI-powered interfaces that enable natural interaction; think voice, not dropdowns

✅ Rapid deployment measured in weeks

✅ Scalable without complexity; connect thousands of machines without lengthy integrations

The companies winning today aren’t planning more; they’re executing faster and adapting continuously.

In this episode of the AI in Manufacturing podcast, I speak with Mickey Shaposhnik, Founder and CEO of Next Plus, about how practical, AI-powered frontline execution is redefining operational excellence.

Watch/Listen now

## Episode: How to Reduce Waste and Improve Efficiency with AI-Powered Quality Control

**Podcast Name:** AI in Manufacturing Podcast (Industry 4.0 TV)

**Episode Title:** How to Reduce Waste and Improve Efficiency with AI-Powered Quality Control

**Guest:** Willem Klein, CEO & Co-Founder, Zetamotion

**Host:** Kudzai Manditereza

**Target Audience:** Manufacturing data leaders, IT/OT solution architects, quality control professionals, and digital transformation leaders implementing AI in industrial operations

---

## 1. Episode Summary

This episode explores how AI-powered quality control can reduce waste and improve efficiency in manufacturing, featuring Willem Klein, CEO and co-founder of Zetamotion. Willem shares why over 90% of industrial AI pilots fail and explains that the real competitive advantage lies not in building bigger AI models, but in designing better end-to-end systems that integrate seamlessly into existing production environments. He introduces Zelia, Zetamotion's AI-powered inspection assistant that reduces model training from weeks of manual data labeling to under an hour using synthetic data and as few as five sample images. The conversation covers the tension between governance and grassroots innovation ("shadow AI"), why manufacturers overwhelmingly prefer edge deployment for quality control data, and why scaling AI across plants is far harder than leadership expects. Willem also shares his vision for fully autonomous inspection systems that configure both software and hardware. Listeners will gain practical insight into what separates successful AI quality control deployments from the 90% that fail.

---

## 2. Key Questions Answered in This Episode

- Why do over 90% of industrial AI pilots fail, and what do the successful ones have in common?

- What is the difference between a model-centric and system-level approach to AI quality control?

- How can manufacturers deploy AI-powered visual inspection without needing an in-house data science team?

- What is synthetic data, and how does it reduce the time and cost of training machine vision models?

- How should manufacturing leaders balance AI governance with grassroots innovation on the shop floor?

- Why do manufacturers prefer edge deployment over cloud for AI-based quality control?

- What makes scaling AI quality control across multiple plants and production lines so difficult?

It requires a strategic architectural approach that effectively utilizes your data ecosystem to make informed decisions in real-time and drive innovation at every level of your organization.

While there are various architectural approaches, the Unified Namespace (UNS) stands out as the most effective approach to achieve optimal results and reap the full economic benefits of digital transformation.

The UNS serves as a critical building block for your overall strategy.

I had the pleasure of interviewing Walker Reynolds, the President of 4.0 Solutions, a renowned industrial IoT expert who introduced and popularized UNS. He provides a masterclass on understanding, implementing, and benefitting from the Unified Namespace approach.

If you'd like to gain a solid understanding of leveraging the Unified Namespace for your manufacturing enterprise, don't miss out on this valuable conversation. 

Outline ✅ Origins and evolution of Unified Namespace ✅ A Description of the Unified Namespace ✅ Why MQTT is the de-facto protocol for UNS implementation ✅ The role of MQTT Sparkplug in UNS Implementation ✅ The role of OPC UA in UNS Implementation ✅ Workflow for designing a Unified Namespace for IIoT System ✅ Mapping physical assets and devices to a Unified Namespace, tools and techniques. ✅ Metadata definition for consistency and accuracy across different systems and devices in UNS ✅ How MES fits into UNS architecture, specific functions and capabilities ✅ Challenges and Mitigation strategies in implementing a Unified Namespace ✅ Impact of Industry4.0 Community Discord Platform ✅ Industry4.0 Influencer Lists ✅ Impact of ChatGPT on Digital Transformation

#iot #iiot #industry40 #industrialiot #uns

## Episode Details

- **Podcast Name:** AI in Manufacturing Podcast (Industry40.tv)

- **Episode Title:** Unlocking Productivity With Casual Models and Agentic AI in Manufacturing

- **Host:** Kudzai Manditereza

- **Guest:** Michael Carroll

- **Guest Title/Role:** Strategic Advisor & Fellow COO Council at LNS Research; Chief Strategy Officer at Trek AI

- **Target Audience:** Manufacturing data leaders, COOs, VP of Operations, IT/OT solution architects, and digital transformation professionals

---

## 1. EPISODE SUMMARY

Agentic AI is not another digital tool to add to the manufacturing technology stack — it is a fundamentally different species of software that treats decisions, not transactions, as the atomic unit of work. In this episode, Michael Carroll, Strategic Advisor at LNS Research and Chief Strategy Officer at Trek AI, explains why US manufacturing productivity has been flat since 2010 despite massive investments in digital tools, and why agentic AI with causal reasoning represents the structural fix. Carroll draws on his 15 years leading digital transformation at Georgia Pacific to reveal how the real productivity killer is not a lack of data or technology, but a cognitive overload crisis combined with organizational permission bottlenecks that drain value from companies in real time. He introduces a practical diagnostic framework — mapping inferencing load and permission load — that any operations leader can apply today to identify where value is leaking from their organization and where agentic AI can deliver immediate impact.

---

## 2. KEY QUESTIONS ANSWERED IN THIS EPISODE

- Why has US manufacturing productivity been flat since 2010 despite massive digital investments?

- What is agentic AI, and how is it fundamentally different from traditional manufacturing software like MES and ERP?

- What is causal reasoning, and why does it matter more than explainable AI for manufacturing decisions?

- How does the permission architecture in manufacturing organizations destroy value and slow decision velocity?

- Where should COOs and VPs of Operations start when preparing their organizations for agentic AI?

- Why do alignment meetings signal that a company's numbers can't be trusted?

- How should IT and OT organizations restructure their relationship to enable competitive advantage?

---

Without strong connectivity, AI remains underutilized, limited by data silos, and inconsistent integration.

Connectivity isn’t just about linking devices; it’s about creating a unified data environment where AI can operate at its full potential—powering everything from predictive maintenance to automated quality control and beyond.

To learn more about IT/OT connectivity for enabling AI use cases in manufacturing I had a conversation with Bernd Hafenrichter who is the CTO of soffico GmbH.

It's a living layer of knowledge built through problem-solving, conversation, and understanding the complex relationships on the factory floor.

This simple truth is often overlooked in industrial data strategies. 

We’ve been conditioned to believe that context can be predefined; baked into standards, taxonomies, and hierarchies.

But in real-world manufacturing, things change, people think differently, and use cases evolve.

So how can we build this dynamic layer of understanding for industrial AI?

In our latest AI in Manufacturing episode, I spoke with Bob van de Kuilen, Director at Thred, about a more human-centric approach to industrial data contextualisation using Knowledge Graphs.

Thred is a tool that plugs into Ignition Platform, enabling users to visualize their factory assets in a knowledge graph, link related data points, embed domain expertise, and deliver structured, contextualized data to AI and analytics tools.

We discuss:

 ✅ Why traditional approaches to data context often fail

 ✅ Knowledge Graphs act as a mind map for data

 ✅ The practical steps to building context

 ✅ How this new context layer serves as the perfect foundation for AI agents.

✅ Drivers for Data-Driven Manufacturing ✅ Introducing MES for Manufacturing ✅ Functions of Manufacturing Execution Systems ✅ Selecting and Implementing an MES System ✅ Modern Manufacturing Execution System Architecture ✅ Problems with Current Manufacturing Data Systems ✅ Unified Namespace for Manufacturing Execution System ✅ Brownfield Integration into Unified Namespace Architecture ✅ The Future of MES in Manufacturing

​

Correlation ≠ Causation

Just because two variables move together doesn’t mean one causes the other. 

This blind spot can lead to poor decisions and surface-level fixes that don’t solve the real issue.

For example, a machine’s temperature spikes often coincide with defects. Traditional analytics might alert you when it happens—but not why. Is it the temperature? A faulty sensor? Operator error?

Causal Inference flips the script. Instead of just observing data patterns, it asks:

“What actually caused this outcome?”

I recently sat down with Daniele Gamba, CEO of AISent Srl to learn more about building industrial intelligence solutions with Caussal AI.

Here are the key takeaways:

The Shift from Centralized to Agentic AI Manufacturers can no longer afford to rely on centralized data science teams that create bottlenecks. Agentic AI empowers subject matter experts (SMEs) to directly access and analyze data, significantly cutting down response times from weeks to minutes.

A Human-Level Automation Approach Agentic AI mirrors human problem-solving by breaking down complex challenges into smaller tasks. This approach enables real-time, adaptive solutions that evolve with the operational landscape, especially when processes change or equipment gets upgraded.

Beyond Traditional Machine Learning Unlike static models that rely on historical data, self-adaptive systems continuously learn in real-time. This makes AI solutions more relevant to the current operational environment, minimizing the need for human intervention.

Real-World Applications and Results Youssf shared compelling use cases where agentic AI is deployed for quality monitoring, helping companies address shifts in product metrics faster and more effectively. Whether it's vibration analytics or yield optimization, the results are astonishing.

Here are five key takeaways:

In the latest episode of the AI in Manufacturing podcast, I sit down with Jeff Knepper, President at Flow Software Inc., to discuss the intersection of Information Management and AI in modern manufacturing, plus the exciting announcement of Timebase Atlas launch.

Here’s some of what we cover in this episode:

✅ Why manufacturers struggle to make use of their data

✅ Building reliable pipelines for AI-driven use cases

✅ AI Agents in Manufacturing – Where they fit and what they need

✅ Unified Analytics Framework vs. Unified Namespace

✅ Historization Strategies – Best practices from edge to cloud

✅ Timebase Atlas Launch Announcement: Data Modeling, Pipelines, Knowledge Graphs, and AI interfaces

✅ MCP and Flow AI Gateway: Beyond APIs to Context-Aware Agent Interfaces

Outline ✅ What is DataOps, and Why is it relevant for Digital Transformation? ✅Key Steps of Implementing DataOps for a digital transformation strategy. ✅ The Role of DataOps in a Unified Namespace Architecture ✅Typical Data Sources for Integrating into DataOps Pipeline ✅ Effective practices for modeling, normalization, and contextualization of Industrial data. ✅ Data Modeling Standards for DataOps, Pros, and Cons ✅The role of MQTT and Sparkplug in a DataOps-based Strategy ✅ The role of OPC UA in DataOps Strategy ✅ DataOps for Historical Data in the Unified Namespace ✅ REST API for real-time and transactional data harmonization ✅ Best practices for DataOps deployment

I speak with Melvin Francis, the Project Manager for OPC UA and TSN-related services at BE.services GmbH, to discuss the challenges faced in developing related applications due to the lack of ready-made OPC UA + TSN stacks or SDKs.

We also delve into the value of OPC UA + TSN integration and how it can address the requirements of time-critical industrial systems.

Furthermore, we explore the ongoing research project by BE Services, in partnership with Hochschule Offsberg University, the German government, and various hardware vendors, to develop a hardware and OS-independent low-footprint OPC UA + TSN SDK for wired and wireless TSN networks.

Below is the outline of our conversation.

✅ How Time Sensitive Networking (TSN) Works ✅ OPC UA PubSub + TSN Integration ✅ Low-footprint OPC UA PubSub TSN Architecture ✅ SDK Test Suite for low-cost testing ecosystem ✅ Practical Demonstration

In our latest podcast episode, I had the pleasure of speaking with Cyrus Shaoul, CEO of Leela AI, about visual intelligence and its transformative impact on manufacturing operations. 

Here are some Key Takeaways:

1️⃣ Beyond Traditional Machine Vision

Unlike traditional machine vision systems that focus on product inspection, visual intelligence looks at the entire manufacturing process. It helps identify value-adding activities in real-time, ensuring operational excellence is met consistently.

2️⃣ Uncover Hidden Performance Insights

By integrating visual intelligence, companies can detect bottlenecks and wasted time during manual operations. In one case, Lila AI improved line capacity by 20% by identifying areas where standard operating procedures weren’t being followed.

3️⃣ Boost Safety & Compliance

With advanced monitoring, manufacturers can significantly reduce safety violations. One customer saw a 50% reduction in non-compliant events, leading to fewer accidents and a safer work environment.

4️⃣ Improving Quality Control

Visual intelligence doesn’t just ensure processes run smoothly; it improves quality control by catching invisible defects in real-time, boosting yields by 10%. This kind of proactive monitoring helps prevent costly mistakes that traditional methods might miss.

5️⃣ Faster, Data-Driven Decisions

With visual intelligence, data is constantly collected and analyzed, allowing teams to make real-time adjustments and enhance productivity, safety, and quality simultaneously. The ROI on this technology speaks for itself.

🎧 Tune in to hear the full conversation and explore how visual intelligence is reshaping the future of manufacturing.

With the capacity to locally integrate industrial data and transfer it via a private LoRaWAN network over vast distances, LoRaWAN simplifies protocol conversion and enhances data recovery.

To learn more about the LoRaWAN for Industrial IoT applications I had a chat with Wienke Giezeman. Wienke is the CEO & Co-founder at The Things Industries, a scalable LoRaWAN solutions provider, and the initiator of The Things Network, the first crowdsourced free and open 'Internet of Things'

Here's the outline of our conversation.

✅ Introduction to LoRaWAN and The Things Stack ✅ Common LoRaWAN Use Cases in Industrial IoT. ✅ Typical architecture of a LoRaWAN solution for IIoT. What are the key edge components involved? ✅ Key considerations for designing a LoRaWAN wireless sensor network for IoT. ✅ Best practices for Integrating LoRaWAN data into IIoT Platforms and Controls Networks. ✅ LoRaWAN gateway and node selection. ✅ Security in LoRaWAN networks ✅ The role of MQTT and other IIoT protocols in LoRaWAN deployments. ✅ Enterprise and cloud integration capabilities of The Things Stack ✅ Emerging trends or features in the LoRaWAN space.

Inefficiencies in Equipment Usage: Downtime, slow cycle times, and unidentified bottlenecks reduce Overall Equipment Effectiveness (OEE), leading to wasted resources and missed production targets.

Safety Risks: Ensuring worker safety while maintaining productivity is difficult, especially in environments with heavy machinery and fast-moving processes.

Despite best efforts, traditional methods struggle to keep up with the complexity and speed of modern manufacturing.

By using computer vision and deep learning, Video AI Agents bring continuous, detection and response of issues—far beyond what traditional methods alone can achieve.

I recently sat down with Karim Saleh, Co-founder and CEO at Cerrion to learn more about how to Maximize OEE and production Line Safety with Video AI Agents

As CEO of Cybus, he has been advising and guiding global enterprises towards decentralized, secure Smart Factory and data-driven Smart Services across various industries such as automotive and battery manufacturing, machinery and tool builders or metal processing.

Outline Introduction to Infrastructure as Code for Industrial Digitalisation IaC workflow for streamlining the deployment of industrial digital solutions IaC for management of industrial software configuration  Balance between UI and DevOps for OT Engineers What does High Availability and Scalability mean for OT? Effective data governance strategy for digital transformation? Cybus Connectware IaC architectural layout Azure IoT operations vs HiveMQ and Cybus IaC Case Studies The Future of IaC

As a result, manufacturers normally place barcodes after key part transformations.

That means, for 70%+ of the production process, you're flying blind. You're guessing which parts went through which treatments.

And when something fails? You're looking at massive recalls, supplier penalties, and lost time.

What if we could code physical parts in a way that never fades?

That’s exactly what Serra Tuzcuoglu and her Co-Founder invented. CDOT AI Code, a frequency-based, AI-readable identifier that can survive the harshest industrial conditions.

Unlike traditional visual patterns, it uses signal recognition instead of contrast, enabling it to be read even when surfaces are scratched, painted, or distorted.

A global appliance manufacturer gave them a challenge: “If your code can survive enamelling and high heat, we’ll use it.”

It did. That was the beginning. CDOT AI Code Code is now deployed globally, from Renault crankshafts to Ford EV battery trays, military tanks to printing cylinders.

In the latest episode of the AI in Manufacturing podcast, I sat down with Serra Tuzcuoglu, the CEO and Co-Founder of COSMODOT, to discuss:

✅ How CDOT AI Code works

✅ Solution Architecture & Integration into the factory network

✅ Readiness for AI-based quality analytics and creation of digital twins.

✅ Real-world examples and Case Studies

And yet, its remarkable potential remains largely untapped in the industrial software domain.

Having been a Node-Red user and promoter since its early days, I've observed that its preparedness for enterprise deployments is a key challenge limiting its broader adoption.

To discover the most effective strategies for deploying and managing enterprise-grade Node-Red applications in Industrial IoT, I invited Nick O'Leary for a podcast conversation.

As the co-creator of Node-Red and the CTO and Founder of FlowForge—a DevOps platform for Node-Red—Nick brought invaluable insights to the discussion.

Below is the outline of our discussion.

​

✅ Origins and Evolution of Node-Red

✅ Challenges and best practices for deploying Node-RED in Industrial environments

✅ DevOps for Node-RED in IIoT

✅ Key Features of Enterprise Ready Node-RED

✅ Best practices for industrial data acquisition using Node-Red

✅ Suitability of Node-Red for Semantic Modelling

✅ Node-Red vs Off-Shelf Solutions

✅ Use cases for Edge vs Cloud deployment of Node-Red

✅ Best Practices for Scaling Node-Red in large-scale IIoT deployments.

✅ Access control, data encryption, and authentication in Node-Red.

✅ Examples of successful IIoT projects using Node-Red

✅ The Future of Node-Red in Industrial IoT Applications

​

Here's what we discussed

✔️ What is a digital twin? ✔️ Digital modeling concepts ✔️ Types of digital twins ✔️ Digital Twin formats ✔️ Tools and frameworks ✔️ Step by Step Process ✔️ Interfacing with digital twins ✔️ Digital Twin vs Digital Thread ✔️ AI and ML in Digital Twins ✔️ Digital Twins and Blockchain ✔️ Digital Twin Consortium Pieter is the CEO of XMPRO, a low-code application development platform that enables subject matter experts to build and deploy real-time applications in weeks. He currently serves as the co-chair of the Natural Resources Working Group for the Digital Twin Consortium & previously co-chaired the Digital Twin Interoperability, Industrial Digital Transformation, & Distributed Ledger Task Groups at the Industrial Internet Consortium (IIC)

1️⃣ Labor Challenges Meet Smart Solutions

• Manufacturers face critical labor shortages, resulting in significant costs. Zeeshan shared how AI-powered Assembly Co-Pilots are slashing error rates and scrap costs by up to 90% while empowering workers with real-time guidance.

2️⃣ Merging Lean Principles with AI

• Traditional lean manufacturing focuses on quality, productivity, and safety. RetroCausal’s tools like Kaizen Co-Pilot and Ergo Co-Pilot seamlessly integrate lean methodologies with advanced AI, accelerating time studies and ergonomic assessments in hours instead of weeks.

3️⃣ Scalability Across Diverse Workflows

• From discrete manufacturing to medical devices, AI co-pilots are not just for single processes—they scale efficiently across multiple sites, even in highly regulated industries.

That's why I'm excited about Tulip Interfaces' new "Frontline Copilot" which uses LLMs for natural language interaction between operators and manufacturing systems.

To truly comprehend its significance and the broader implications of Applied AI in manufacturing, I spoke with Roey Mechrez, PhD in my latest podcast episode.

Roey is the Head of AI and EMEA MD at Tulip Interfaces where he is overseeing Tulip's Machine Learning and Computer Vision strategy.

Here's the outline of our conversation:

Outline ✅ Introduction to the Tulip Ecosystem ✅ Composable, App-based solutions vs. Monolithic MES ✅ Tulip for Process Engineer, Operator, Manager End Users ✅ Technology stack for Modern Manufacturing ✅ Tulip Ecosystem for Applied AI in manufacturing ✅ Connecting shop-floor visuals to advanced analytics tools. ✅ How Data is Shaping the Next Layer of Manufacturing ✅ Tulip connectivity and AI integrations for building shop-floor solutions ✅ Introducing Tulip Frontline Copilot: Why Generative AI Matters for Manufacturing ✅ Natural Language Operator Interaction with Tulip Copilot, Use Cases ✅ Integrating legacy machine data into modern systems with Machine Learning and Edge Connectivity ✅ Computer Vision Capabilities and Third-party Integrations in the Tulip Ecosystem. ✅ Driving Forces for AI Adoption in Manufacturing ✅ The Future of AI in Manufacturing

Here are some key takeaways:

82% of Failures Are Random Nikunj explains that a staggering 82% of failures in industrial systems appear random. Without understanding their causes, manufacturers struggle to prevent them. This is where smart, data-driven actions come into play to improve decision-making and reduce failures​.

Condition-Based Actions: A Game Changer In manufacturing, decisions often rely on experience, which can take years to accumulate. Condition-based actions allow manufacturers to make smarter decisions without needing decades of experience. By detecting and acting on real-time conditions, manufacturers can optimize maintenance, improve quality, and reduce emissions​.

Real-Time Data = Real-Time Decisions From mining to steel production, the power of real-time data can revolutionize how we handle variations in materials, weather conditions, and equipment performance. Nikunj shares how timely insights enable proactive decision-making, reducing downtime and energy waste​.

Smart Guidance Systems Smart manufacturing requires systems that can analyze data in real-time and offer actionable guidance. Think of it like a GPS for your factory: these systems navigate complex production challenges and direct optimal actions for maintenance, quality control, and emissions​.

What's Next for Smart Manufacturing? Nikunj forecasts that the next step in manufacturing will be integrating smart guidance systems across various processes—from maintenance to quality assurance—allowing companies to move from reactive to proactive management​.

## Episode: Designing Autonomous AI Agents for Industrial Operations

**Podcast Name:** AI in Manufacturing Podcast (Industry 40.tv)

**Episode Title:** Designing Autonomous AI Agents for Industrial Operations

**Guest:** Kence Anderson, CEO & Founder, AMESA

**Host:** Kudzai Manditereza

---

## Episode Summary

This episode explores how autonomous AI agents can transform industrial operations through a methodology called machine teaching. Kence Anderson, CEO and founder of AMESA, draws on eight years of experience applying autonomous systems to manufacturing and logistics to explain why more than 95% of what's called "industrial AI" today is really just data storage and connectivity — missing the actual intelligence layer that can perceive and act. Anderson breaks down his machine teaching methodology, which captures expert operator knowledge and structures it into teams of specialized AI agents that learn by practicing in simulation before deploying to the factory floor. The conversation covers multi-agent design patterns, the AMESA platform's three core products (Agent Orchestration Studio, Agent Cloud, and Runtime), and real-world examples from Fortune 500 glass manufacturers, beverage companies, and logistics operations. Listeners will learn why monolithic AI approaches fail in manufacturing, how to avoid pilot purgatory, and how companies can go from data to deployed autonomous agents in approximately 12 weeks.

---

## Key Questions Answered in This Episode

- What is machine teaching and how does it differ from traditional machine learning approaches in manufacturing?

- Why has manufacturing productivity remained stagnant despite massive investments in IoT and data infrastructure?

- What are the four fundamental ways AI systems can make decisions in industrial environments?

- How do multi-agent design patterns work for industrial automation, and why do they outperform monolithic AI?

- What does it take to scale AI agents across multiple plants, production lines, or product recipes?

- How do you bridge the gap between AI training in simulation and real-world deployment on legacy factory systems?

- What is pilot purgatory and how can manufacturers avoid it when implementing industrial AI?

---

We discussed leveraging data for efficiency, the challenges of data quality, and choosing between foundational principles and cutting-edge ML algorithms.

Jim also highlighted the significance of tools and strategies in this sphere, emphasizing the urgency of digitizing domain knowledge in the face of an impending knowledge drain.

Jim, is the President and Founder of Industrial Insight, Inc. where he helps industrial companies turn data into actionable information to deliver tangible results for their organization.

Here is the outline of our conversation:

✅ Principles of Data-Driven Process Optimization ✅ Opportunities in data-driven optimization and use case ✅ Challenges faced by industries when implementing data-driven optimization strategies? ✅ Overcoming the hurdles of data quality and fidelity? ✅ First principles vs. Multivariate data analysis vs. ML algorithms? ✅ Evaluating readiness to effectively integrate AI/ML in process optimization ✅ Tech stack for data-driven optimization ✅ Impending knowledge drain, and capturing domain knowledge into digital tools.

## Episode: Scaling Industrial Intelligence with the I3X Common API

**Podcast Name:** AI in Manufacturing Podcast (Industry40.tv)

**Episode Title:** Scaling Industrial Intelligence with the I3X Common API

**Guest Name:** Matthew Parris

**Guest Title/Role:** Director of Quality Test Systems, GE Appliances; Leading Contributor to the I3X Specification

**Host:** Kudzai Manditereza

---

## 1. Episode Summary

This episode explores how the Industrial Information Interoperability Exchange (I3X) common API is poised to become the universal interface for accessing manufacturing data across software platforms. Matthew Paris, Director of Quality Test Systems at GE Appliances and a leading contributor to the I3X specification, explains why the manufacturing industry has lacked a standardized way to retrieve information from Level 3 and Level 4 software systems — and how I3X solves this by leveraging simple, proven IT technologies: HTTP and JSON. Paris draws a compelling analogy between I3X and the early web browser revolution, comparing the I3X Explorer tool to Netscape's role in breaking down walled-garden internet portals. The conversation covers how I3X differs from OPC UA and MQTT, why a vanilla MQTT broker is insufficient for a true Unified Namespace, and how standardized interfaces accelerate AI deployment in manufacturing. Listeners will gain a clear understanding of where I3X fits in modern industrial architectures and why now is the time to get involved with the specification while it's in beta.

---

## 2. Key Questions Answered in This Episode

- What is I3X and what problem does it solve for manufacturers?

- How is I3X different from OPC UA and MQTT?

- Why is an MQTT broker alone not sufficient for a Unified Namespace (UNS)?

- How does I3X enable manufacturers to scale from data visibility to operational AI?

- Where does I3X fit in a modern industrial architecture alongside UNS and MQTT brokers?

- Why does I3X support OPC UA Part 5 information models, and how should manufacturers think about data typing?

- How will I3X achieve vendor adoption without a chicken-and-egg problem?

## Episode: Optimizing AI Inference for Agentic Operations in Manufacturing

**Podcast Name:** AI in Manufacturing Podcast (Industry40.tv)

**Episode Title:** Optimizing AI Inference for Agentic Operations in Manufacturing

**Guest:** Kelvin Cooper, Co-Founder & CEO, Neurometric.ai

**Host:** Kudzai Manditereza

---

## 1. Episode Summary

This episode explores why manufacturing companies struggle to scale AI from pilot to production—and how inference orchestration and small language models (SLMs) offer a practical path forward. Kelvin Cooper, Co-Founder and CEO of Neurometric.ai, joins host Kudzai Manditereza to break down why routing all AI tasks through a single frontier model becomes a cost and reliability liability at scale. Cooper draws on his background in venture capital, private equity AI rollups at Pilot Wave Holdings, and AI policy research at the Milken Institute to argue that the future of industrial AI is not one model that knows everything, but a coordinated system of specialized models that each know their job. The conversation covers Neurometric's AI maturity framework, real customer results showing 10x cost and latency improvements, the concept of catastrophic forgetting, and why manufacturing leaders need to adopt a startup execution mindset rather than over-analyzing use cases. Leaders seeking to cut AI inference costs and accelerate deployment will find actionable strategies throughout.

---

## 2. Key Questions Answered in This Episode

- Why do 95% of AI proof-of-concepts in manufacturing never make it to production?

- How should manufacturers select their first AI use case instead of getting stuck in analysis paralysis?

- What is inference orchestration and why does it matter for scaling AI in manufacturing?

- Why is relying on a single large language model a liability for industrial AI at scale?

- What are small language models (SLMs) and how do they deliver faster, cheaper, and more accurate AI?

- What is catastrophic forgetting and how does it affect AI deployments in manufacturing?

- How can manufacturers avoid vendor lock-in when building AI systems?

---

**Podcast Name:** AI in Manufacturing Podcast 

**Episode Title:** How to Build AI Solutions That Actually Work on the Factory Floor

**Guest:** Renan De Villiers, Founder & CEO, OSS Ventures

**Host:** Kudzai Manditereza

---

## 1. Episode Summary

This episode explores why only 5% of factories currently operate like tech companies — and what it will take to reach 50% within a decade. Renan De Villiers, founder and CEO of OSS Ventures, a Paris- and Boston-based venture builder with 22 spun-out companies live in 3,800 factories worldwide, shares hard-won lessons from visiting over 900 manufacturing sites and deploying AI across 100+ factories in the past two years. Drawing on his background as a former McKinsey consultant, factory director, and tech startup founder, De Villiers explains why most manufacturing AI initiatives fail, how to industrialize the discovery process, and why designing the human experience of managing AI agents is the most underestimated challenge in scaling industrial AI. Listeners will learn the concrete frameworks OSS Ventures uses to validate problems before building, the "10x test" for deciding what to pursue, and why the factory of the future requires fewer but far better-paid people. This episode is essential for anyone leading AI adoption in manufacturing or building software products for the factory floor.

---

## 2. Key Questions Answered in This Episode

- **What does a tech-enabled factory look like compared to a traditional factory?**

- **Why do 85% of manufacturing AI projects fail, and how can you beat those odds?**

- **How do you identify the right AI use cases on the factory floor?**

- **What is the "10x test" for validating manufacturing AI opportunities?**

- **Why is tribal knowledge the biggest hidden barrier to AI in manufacturing?**

- **How do you scale an AI solution from one factory to hundreds?**

- **Should AI be embedded into existing products or built as a new experience layer?**

---

---

## 1. Episode Summary

This episode explores how agentic AI is creating a new category of digital skilled workers for manufacturing, addressing the industry's deepening productivity plateau and skilled labor crisis. James Zheng, Co-Founder and Chief Product Officer of Optimate AI, draws on over a decade of experience building and deploying manufacturing software — from SAP's cloud ERP to PTC's ThingWorx IoT platform — to explain why traditional digital transformation investments have failed to move the productivity needle. Zheng introduces the concept of a "decision intelligence and execution layer" that sits on top of existing systems of record (MES, ERP, CMMS, SCADA) to orchestrate AI agents that augment engineers, technicians, and frontline leaders. The conversation covers practical adoption patterns, the critical role of knowledge graphs and context graphs, why perfect data isn't a prerequisite for getting started, and real-world use cases in automotive and discrete manufacturing. Listeners will walk away with a clear framework for identifying, prioritizing, and scaling agentic AI use cases on the shop floor.

---

## 2. Key Questions Answered in This Episode

- What is agentic AI and why should manufacturers care about it now? - What is the skilled labor crisis in manufacturing and how does agentic AI address it? - What is the difference between a knowledge graph and a context graph in industrial AI? - How should manufacturers approach data readiness for AI agent deployment — do you need perfect data? - What are the best first use cases for AI agents on the factory floor? - How does a decision intelligence layer differ from adding a copilot to existing manufacturing software? - How should manufacturing leaders balance top-down AI governance with bottom-up frontline innovation?

**Podcast Name:** AI in Manufacturing Podcast 

**Episode Title:** How to Build and Scale Agentic AI Workflows in Manufacturing

**Guest:** Bernard Kraswald, Co-Founder & CTO at Ethon AI

**Host:** Kudzai Manditereza

---

## Episode Summary

This episode explores how manufacturers can build and scale agentic AI workflows to achieve operational excellence across factories. Bernard Kraswald, Co-Founder and CTO at Ethon AI, explains why traditional continuous improvement methods have reached their limits and how purpose-built industrial AI—grounded in process knowledge graphs and causal reasoning—unlocks the next wave of manufacturing optimization. Key insights include why deep data contextualization through knowledge graphs is essential for agentic AI (not just basic tag hierarchies), how causal AI differs from correlation-based analytics by making root cause findings actionable, and why a layered architecture of data infrastructure, specialized model layer, and application layer prevents hallucinated recommendations in safety-critical environments. Bernard also shares real-world results, including a globally scaled deployment at Siemens that generated over $10 million in documented savings. Whether you're evaluating industrial AI platforms or architecting your data stack for agentic workflows, this episode provides a practical roadmap from data ingestion to autonomous process control.

---

## Key Questions Answered in This Episode

- What is a process knowledge graph, and why is it essential for agentic AI in manufacturing?

- How does causal AI differ from correlation-based analytics in industrial settings?

- What architecture layers are needed to run agentic AI workflows reliably in manufacturing?

- Why can't general-purpose LLMs like ChatGPT or Claude replace purpose-built industrial AI models?

- How do you build a knowledge graph iteratively without delaying ROI?

- What does a typical deployment timeline look like for industrial AI platforms?

- How should manufacturers handle security and governance when connecting OT systems to cloud-based AI?

---

**Podcast Name:** AI in Manufacturing Podcast 

**Guest:** Walker Reynolds, President & Solutions Architect at 4.0 Solutions, Founder of the PROVE IT Conference

**Host:** Kudzai Manditereza

**Target Audience:** Manufacturing data leaders, IT/OT solution architects, and digital transformation professionals

---

## Episode Summary

Walker Reynolds, President and Solutions Architect at 4.0 Solutions and founder of the PROVE IT conference, delivers an unfiltered assessment of where industrial AI actually stands in 2025. Drawing from conversations with over 1,000 attendees at this year's PROVE IT conference—70% of whom were end users working in manufacturing—Reynolds identifies three critical industry shifts: AI fatigue is setting in as vendors outpace market readiness, knowledge graphs have emerged as the essential technology for enabling agentic AI in manufacturing, and the gap between digitally mature and immature manufacturers is widening. The conversation covers why most manufacturers still aren't getting value from their unified namespace implementations, the five most practical AI applications seen at PROVE IT, and why autonomous agents are a mathematical impossibility given current LLM reliability. Reynolds closes with his complete recommended technology stack for manufacturers and a prediction that plant floors will see *more* people, not fewer—but they'll be analysts supervising AI agents rather than middle managers managing people.

---

## Key Questions Answered in This Episode

- What is the current state of AI adoption in manufacturing in 2025?

- Why are some manufacturers failing to get value from unified namespace implementations?

- What role do knowledge graphs play in enabling agentic AI for manufacturing?

- What are the most practical AI applications for manufacturers right now?

- Can AI agents run autonomously in manufacturing operations?

- What does the ideal industrial data architecture stack look like for a small to midsize manufacturer?

- How does unified namespace serve as the backbone for agentic AI?

---

Episode Title: Context Engineering Techniques for Building Reliable Industrial AI Agents

Guest: Zach Etier, VP of Architecture at Flow Software

Host: Kudzai Manditereza

Episode Summary

This episode explores context engineering — the discipline of curating and managing the information supplied to AI agents — and why it is the key to building reliable industrial AI systems. Zach Etier, VP of Architecture at Flow Software, joins host Kudzai Manditereza to break down why simply pumping more data into an AI agent's context window actually degrades performance through dilution, hallucination, and lost instructions. Zach walks through three core context engineering techniques — persisting context, summarization/compaction, and isolation via sub-agents — and explains how each one maps to real manufacturing use cases like automated shift-handover reports. The conversation also covers the practical differences between skills, MCP servers, and sub-agents, and why deterministic code should handle calculations while agents handle orchestration. Finally, Zach makes the case that knowledge graphs with formal ontologies will become essential data architecture for scaling industrial AI across the enterprise. Whether you are evaluating your first agent pilot or planning multi-site deployment, this episode provides a concrete framework for engineering context that agents can reliably act on.

Key Questions Answered in This Episode

• What is an industrial AI agent, and how does it differ from a chatbot or general-purpose LLM?
• Why does giving an AI agent more context actually reduce its performance?
• What is context engineering, and why is it replacing prompt engineering for agentic AI?
• What are the three core techniques for managing an AI agent's context window in manufacturing?
• How should you decide when to use skills vs. MCP servers vs. sub-agents?
• Why should deterministic code handle calculations instead of letting the AI agent compute them?
• How do knowledge graphs and ontologies enable enterprise-scale industrial AI?

This episode explores why most manufacturing AI initiatives fail and what companies must do to build a foundation for AI-native industrial intelligence. Craig Scott, Founder and CEO of Fuuz, an industrial intelligence platform, shares insights from nearly a decade of bridging the gap between shop floor data and enterprise systems. The conversation reveals why the missing "shim" between operational technology and enterprise systems is the root cause of unreliable data in manufacturing, and why model-driven approaches are essential for scaling AI across industrial operations. Craig explains how organizations can achieve a single source of truth by implementing a persistent contextualization layer that governs data before AI ever touches it. Whether you're struggling with fragmented point solutions, evaluating industrial data platforms, or preparing your data infrastructure for AI, this episode provides a practical framework for building scalable industrial intelligence.

2. KEY QUESTIONS ANSWERED IN THIS EPISODE

• What is fundamentally broken with current manufacturing data infrastructure and how does it impact AI initiatives?
• Why do most AI pilots fail to scale in manufacturing environments?
• What is a model-driven approach to industrial data, and why is it superior to in-line data transformation?
• How do you balance enterprise governance with plant-level flexibility in industrial data architectures?
• Should manufacturers adopt industry-standard data models like ISA-95 or build custom models?
• What is the difference between a data lake and an operational intelligence platform?
• How can manufacturers prepare their data foundation before investing in AI technologies?

However, this process requires OT and IT knowledge not present in many

industrial companies, mainly SMEs.

Ander Garcia Gangoiti and his team built a micro-service edge architecture based on MQTT, TimescaleDB, Node-Red and Grafana stack to ease the integration of soft AI models into industrial system.

The architecture has been successfully validated controlling the vacuum

generation process of an industrial machine.

Soft AI models applied to real-time data of the machine analyze the vacuum value to decide when the most suitable time is:

⇨ to start the second pump of the machine,

⇨ to finish the process, and

⇨ to stop the process due to the detection of humidity.

Ander is the Director of Data Intelligence for Industry at Vicomtech and I recently sat down with him on the AI in Manufacturing podcast.

This is the next chapter of industrial process automation.

Already implemented at the largest Oil refinery in the world, Software-defined control systems break the traditional link between hardware and logic.

This separation allows for dynamic control, centralized intelligence, and flexible deployment across complex industrial environments.

When integrated with time-series foundation models, these systems harness AI for intelligent loop control, advanced process optimization, and even reinforcement learning, driving unprecedented levels of performance in control environments.

In the latest episode of the AI in Manufacturing podcast, I sat down with Huize Zhang to explore this transformation. Huize is the Vice President at SUPCON, China’s leading DCS provider, and the founder of FREEZONEX, an open-source IIoT platform.

Here’s the outline of our conversation:

-The Control Platform of The Future

-Open Standards and Platforms

-AI-Driven Optimization in Process Industries

-Time-Series Pre-Trained Transformers

-Reinforcement Learning in Process Industries

-UNS Integration with AI Agents

Here’s the outline of our conversation:

• Core Challenges in Managing Industrial Data for Data‑Driven Manufacturing
• The Role of Ontologies and Knowledge Graphs in Advancing Industrial Data Interoperability and Analytics
• IIoT Data Integration and Standardization Approaches 
• Semantic‑Modeling Best Practices for Scaling Value Creation
• Using Knowledge Graphs as Infrastructure for Digital Twins and Industrial AI
• Industrial AI Use Cases Powered by Knowledge Graphs
• The Real Business Value of Digital Twins in Manufacturing
• Building the Next-Gen Digital Twins with AI, LLMs, and Knowledge Graphs
• AI Agents, and MCP for Distributed Intelligence on Digital Twins
• Multi-Agent AI Systems for the Future of Manufacturing Digitalization

While traditional inspection methods have been the cornerstone of quality control in manufacturing, they come with limitations such as subjectivity, fatigue, and scalability challenges.

AI-powered visual inspection systems address these issues.

Leveraging advanced algorithms and machine‑learning models, they analyze images with high accuracy, identifying defects that may be invisible to the human eye. 

This not only enhances the reliability of quality assessments but also increases operational efficiency, allowing manufacturers to streamline their processes and reduce costs. 

The capability to detect anomalies in real-time empowers companies to address issues before they escalate, ensuring that only the highest-quality components progress through production.

To find out more about the application of Visual AI Inspection in manufacturing, I recently sat down with Priyansha Bagaria who is the Founder and CEO of Loopr AI. 

Why?

Because the data is too vast, too fast, and too fragmented for any human to handle in real-time.

Even the best engineers can’t monitor thousands of variables 24/7.

And failing to harness this data has real consequences. Critical warning signs of equipment problems or process inefficiencies can be missed, leading to unplanned downtime and quality issues.

The biggest challenge AI Agents solve in industrial enterprises is transforming this overwhelming amount of complex data into actionable intelligence.

However, AI Agents are only powerful for manufacturing data analytics when paired with the right context. 

That means feeding them, sensor data, maintenance logs, ERP & MES records, operator notes, engineering drawings, and SOP documents e.t.c. And quickly surfacing the most relevant information to power rapid AI-driven decision-making.

This is where Vector Storage and Search comes into play.

To learn more about Vector Databases and Data Structure for Industrial AI Agents I had a chat with Humza Akhtar, PhD who is the Senior Industry Principal for Manufacturing and Automotive at MongoDB.

The key lies in addressing foundational challenges and adopting a clear, phased strategy to transform operations.

Fundamentally, AI offers manufacturers a pathway to achieving operational excellence by moving through the four stages of analytics maturity:

1️⃣ Descriptive Analytics – Understanding what happened.

2️⃣ Diagnostic Analytics – Pinpointing root causes.

3️⃣ Predictive Analytics – Forecasting potential equipment failures or quality issues.

4️⃣ Prescriptive Analytics – Recommending the best actions to address challenges.

Despite its promise, many manufacturers struggle with significant obstacles, which include data fragmentation.

I recently had a sit down with Andrew Scheuermann the CEO and Co-Founder of Arch Systems to discuss why building a comprehensive Digital Twin is the key to overcoming these barriers and how manufacturers can use AI to enhance manufacturing workflow efficiency.

Manufacturing presents unique challenges that generic AI solutions often can't effectively address. 

However, by customizing Generative AI systems to meet industry-specific requirements, this gap can be effectively bridged: 

- Tailoring AI to understand specialized language and scenarios enhances its relevance and effectiveness. 

- Integrating additional data sources, such as knowledge graphs, enriches the AI's understanding of relationships and processes unique to manufacturing.

- Implementing safety checks and operational boundaries ensures that AI recommendations are viable, safe, and compliant with industry standards. 

When these measures are in place, Generative AI becomes a powerful tool applicable to a wide range of use cases. 

Tune in to the full episode with Vlad Larichev, the Industrial AI Lead at Accenture Industry X to learn more about Generative AI Use Cases in Engineering and Manufacturing.

Here are some key takeaways:

Federated Learning Enables Cross-Factory Collaboration

Federated learning allows multiple factories to improve AI models without sharing sensitive data. By exchanging learnings, factories can build more robust models while maintaining data privacy and compliance.

Collaboration on Model Training Without Compromising Privacy

One of the biggest challenges in industrial AI is accessing the right data without compromising privacy. Federated learning addresses this by keeping sensitive data local, allowing companies to enhance their AI models collectively without exposing each other’s proprietary or sensitive information.

Cost-Effective Scaling of AI Models Through Reuse

Scaling AI across multiple factories typically involves high costs and complexity. Federated learning significantly reduces development, integration, and operation costs by allowing the reuse of models across different sites without duplicating efforts.

Steamlined Development of Predictive Maintenance and Quality Control Models

Federated learning helps streamline the development of ML models for predictive maintenance and quality control by aggregating insights from multiple sites, reducing the need for extensive data science expertise and making advanced AI accessible to more organizations

Curious about how federated learning can scale industrial AI? Tune in to the full episode to learn more!

This approach, although functional, has been markedly inefficient and reactive, particularly in terms of process optimisation and asset reliability, two vital aspects of industrial operations that can profoundly impact efficiency and profitability.

Digital Twins represent a transformative shift from this reactive approach to a proactive, predictive one. They facilitate a deeper understanding of how systems behave, providing industrial operators with actionable insights that were previously unavailable.

To learn more about the application of digital twins in manufacturing, I had a podcast conversation with Erik Udstuen, who is the CEO and co-founder of TwinThread, a company that provides a digital twin platform that combines Industrial Data with Industrial AI in an integrated development environment for engineers and data scientists.

Here's the outline of our conversation

✅ Challenges driving Digital Twins adoption in modern manufacturing ✅ Key Functions of Digital Twins in Manufacturing ✅ Industrial AI Ops ✅ Use Cases for Asset and Process Digital Twins ✅ Connectivity standards for physical assets to digital twins ✅ Best practices for modelling assets and processes for digital twins  ✅ Effective infrastructure abstraction techniques for Digital Twin Implementation ✅ ISA 88 / 95, and Data Modeling standards for Digital twins ✅ Principles-based vs Machine Learning-based modelling for advanced analytics ✅ Practical examples of successful digital twin applications in Manufacturing  ✅ Selecting a digital twin platform and evaluating capabilities ✅ TwinThread Digital Twin Integrated development environment

We discussed how you can modernize your industrial data architecture to harness the full potential of AI, enhancing both production efficiency and innovation.

Jonathan highlighted three key pillars essential for AI readiness:

Data Accessibility - You can’t train AI without accessible data. Jonathan explains why ensuring your data flows seamlessly across systems is the first critical step.

Data Contextualization - Simply having data isn’t enough. Meaningful, contextualized data is crucial for any AI project to deliver accurate and actionable insights.

Data Relationships - It’s not just about isolated data points. AI thrives on the connections between data points, much like how your operations depend on the synergy between suppliers and internal systems.

Listen to the episode to learn more.

Consequently, implementing a well-thought-out data modelling strategy is key to successful digital transformation as it helps to express the meaning of the data to digital systems.

To learn more about Data modelling for Industrial IoT in general and for Digital Twin use cases in particular, I had a podcast conversation with Erich Barnstedt.

Erich is the Chief Architect for Standards, Consortia and Industrial IoT in the Azure Edge and Platform team at Microsoft

Here's the outline of our conversation

✅ Importance of data modelling for Industrial IoT. ✅ Key Elements of an Effective IIoT Data Model ✅ Standardising Configuration Interface for OPC UA Connectivity Mapping ✅ Manufacturing Ontologies Reference Solution for Digital Twins ✅ UA Cloud Publisher and UA Cloud Twin for Mappping Industrial Assets to Azure Digital Twins using ISA95 ✅ Significance of UA Cloud Commander at the Industrial Edge ✅ OPC UA Information Model Integration using UA Cloud Library ✅ Data Modelling Standards ✅ Web of Things for Endpoint and Interface Description of Industrial Assets. ✅ ChatGPT for fully automating onboarding of non-discoverable industrial assets ✅ Converting proprietary interfaces into OPC UA Information Model using UA Edge Translator. ✅ The role of the IEC/ISO in standardizing data models for IIoT ✅ The Scope OPC UA PubSub Over MQTT in Industrial IoT ✅ Metadata and Type Information in OPC UA PubSub ✅ Industrial Metaverse Reference Architecture with Open Interoperability Standards

​

And for manufacturers, having this intelligent layer at the edge, closer to industrial data sources has proven to be more fitting and valuable.

To gain an understanding of the application of Edge Analytics for intelligent automation, I had a conversation with Martin Thunman. Martin is the CEO and Co-Founder of Crosser, a platform that was built with the realisation that low code edge analytics, automation, and integration software will play a critical role in accelerating the digital transformation journey of Industrial and asset-rich organizations.

You can check out the full conversation on the video linked below.

Outline:

✔️ Challenges in Industrial Legacy System integration with Industry4.0 technologies ✔️ Requirements for next-generation industrial system integration solutions ✔️ Edge Analytics and Opportunities it provides for Industrial Automation ✔️ Functional Composition of an Industrial Edge Analytics Solution ✔️ ISA95 vs Any-to-Any Hub Architecture ✔️ Data Modelling Best Practices for Edge Analytics Solution ✔️ Introduction to Edge Machine Learning Ops ✔️ Required Hardware and OS Capabilities for Edge Analytics ✔️ Crosser - Connectivity to Legacy Industrial Control Systems and Enterprise Applications ✔️ Crosser - Data Management and Orchestration ✔️ Common Uses Cases of Edge Analytics in Manufacturing ✔️ The future of Low Code Platforms in Industrial Automation ✔️ About Crosser

#iot #iiot #industry40 #EdgeAnalytics