The Thinking Factory: How AI Is Giving MES a Brain with Francisco Almada Lobo

What does it take for AI to actually work on the factory floor? Francisco Almada Lobo, CEO and Co-Founder of Critical Manufacturing, joins Phil Seboa and Ed Fuentes to discuss the convergence of MES, IoT, and AI on the shop floor. In this episode: What MES actually is, and how it's evolved beyond process control Why AI pilots in manufacturing don't scale (Audi's 100 initiatives, zero in production) The "fat events" approach: enriching IoT data with MES context for AI workloads Design-time AI vs runtime AI: why deterministic matters Greenfield vs brownfield deployments Configure-to-order manufacturing Medical device digitisation and FDA implications Why this tech revolution is different: there's no time to adjust Book recommendation: "Agentic Artificial Intelligence" by Pascal Bonnet This episode is proudly made possible by FlowFuse. The industrial application platform that helps operations, maintenance, and IT teams connect any machine, move data between systems, and scale industrial applications across sites. Built on Node-RED, with the security and governance you need for real production environments. Learn more at flowfuse.com. Connect with Francisco on LinkedIn: linkedin.com/in/falmadalobo Connect with Phil on LinkedIn: linkedin.com/in/philseboa Connect with Ed on LinkedIn: linkedin.com/in/edfuentes Learn more about Critical Manufacturing: criticalmanufacturing.com Subscribe for more conversations with the practitioners building the future of industrial technology. Website: unpluggediiot.com #Manufacturing #MES #AI #IIoT #Industry40 #podcast Timestamp Chapter 00:00 Introduction 01:37 Francisco's background and founding Critical Manufacturing 06:16 Semiconductor sophistication vs other industries 08:21 "Technology is ready, but factories are not" 09:09 What is MES? Four pillars explained 13:40 MES deployment reality: 3-month MVP to 5-year journey 16:49 "Is an MES ever done?" Continuous improvement cycle 19:52 Greenfield vs brownfield: which is better? 23:40 Build vs buy, and how AI changes the equation 29:04 AI in manufacturing: what's real and what's hype 31:39 Audi's 100 AI initiatives, none scaling to production 35:28 Why manufacturing AI needs time and architecture 37:46 Technology investment cycles shrinking from 30 years to under 1 39:18 The secret sauce: enriching IoT data with MES context 43:55 Fat events, graph databases, and optimising AI workloads 45:42 Low-code + AI: generating metadata, not applications 46:41 Customer stories: semiconductor, configure-to-order, medical devices 50:29 Medical device manufacturing: from paper warehouses to digital 53:07 FDA warning letter for unauthorised AI use 54:11 What Francisco is learning: unprecedented pace of change 57:56 Book recommendation: "Agentic Artificial Intelligence" by Pascal Bonnet 59:19 How to connect with Francisco