Abelara Ascent

Most manufacturers we talk to aren't asking for more software. They're asking for less.Forty applications. A dozen integrations holding it all together.We cover:The supply chain ↔ factory floor disconnect, and why the language of customer POs almost never lines up with the work orders that actually get runWhy standard cost variances catch you at quarter close instead of when you can still do something about themWhat a C-suite KPI dashboard really looks like under the hood — and how many hours of manual reconciliation hide behind every "give me the top three reasons we missed."Craig Scott (CEO, Fuuz) on why this is an architecture problem, not a technology one — and what a shared operational model actually isDylan DuFresne on Fuuz as "IT software that understands OT," Litmus's 250+ device drivers, and modeling data as close to the edge as possibleHussein (Litmus) on edge containerization, metadata at scale, and the UNS flexibility you need to actually move fastThe security question — role-based access control, encrypted data, the Fuuz Gateway through the DMZ, and what "MCP-ready" actually means for OT dataThe brownfield reality — no rip-and-replace, just better connection into the SCADAs, historians, and OPC servers already in placeGlenn's mic drop — if you built top-down from the C-suite, you'd build Fuuz. If you built bottom-up from the factory floor, you'd build Litmus. They're the two most important anchor points in the whole industrial software ecosystem.CHAPTERS 00:00 Intro — the simplification thesis 00:51 Glenn's pain point #1 — supply chain ↔ factory floor disconnect 03:02 Pain point #2 — late awareness of cost variances 04:45 Context — why enterprises are cutting logos 05:48 Pain point #3 — C-suite KPIs vs. how they're really built 07:40 Cappy Hour reference (see the Prove It session) 09:22 The wish — one platform to solve all three 10:16 Craig Scott — it's an architecture problem, not a technology one 11:31 Excel is still the dominant MES 12:27 Shared operational model — red, blue, and purple data 14:09 Walker Reynolds, UNS, and an oil-and-gas project with 50+ software vendors 15:09 Death by a thousand cuts 16:08 Cost visibility isn't a reporting problem — it's a latency problem 17:36 Burying A-players in spreadsheets 18:45 Bridging C-suite KPIs to the plant floor 20:46 The operational model as DNA 22:35 Dylan — getting data from the edge, the right way 25:22 Litmus's 250+ drivers and edge data modeling 27:48 Kepware vs. Litmus — containerization at the enterprise 29:25 Edge modeling as the differentiator 30:48 PLC-level hygiene and OEM involvement 33:40 Patrick's question — security, RBAC, and the Fuuz Gateway 35:46 Litmus + Fuuz visualization — overlap and complement 38:36 Brownfield — Litmus and Fuuz alongside existing OPC, SCADA, historians 43:02 Why point-to-point solutions break at scale 44:38 Glenn's mic drop — top-down vs. bottom-up 46:15 The Lighthouse Partner — how to work with Abelara 47:19 Closing — we always start from a fresh sheet of paper\WORK WITH ABELARA — LIGHTHOUSE PARTNER PROGRAMCompanies come to Abelara to architect agnostic solutions and skip past the trial-and-error. We're accepting a small number of Lighthouse Partners — a digital transformation workshop, architected solution, and hitting the ground running.

The conversation begins with an introduction to predictive maintenance and the expert panel. Glenn Gardner discusses the evolution of predictive maintenance, highlighting the transformation of the industry over the past 20 years. He then delves into the concept of vibration as a predictive technology, providing a detailed explanation of its relevance and power. The discussion shifts to the role of CMMS in predictive maintenance, emphasizing the importance of both predictive technology and CMMS for an effective maintenance program. Mark Kingkade shares insights on vibration analysis in real systems, showcasing the practical application of vibration analysis in identifying machinery flaws. The conversation covered challenges in equipment repair, integration of machine data with maintenance workflows, data integration and predictive maintenance, and integrating advanced maintenance technologies. The challenges included difficulty in finding replacement parts, extended lead times for replacements, and the impact of supply chain challenges. The integration of machine data with maintenance workflows highlighted the direct integration of diagnostic insights with work orders, automated generation and assignment of work orders, and the importance of getting information to the right people. Data integration and predictive maintenance discussed the integration of machine data from various sources, usage-based predictive maintenance, and a phased approach to implementing predictive maintenance. Integrating advanced maintenance technologies addressed challenges in integrating different maintenance technologies, the importance of keeping analysis in purpose-built tools, and sending scalar data to centralized systems.TakeawaysPredictive maintenance has evolved significantly over the past 20 years, with a notable reduction in cost and time for implementing a world-class program.Vibration analysis is a powerful technology for identifying machinery flaws, and when combined with CMMS, it forms an effective predictive maintenance program. Challenges in equipment repair due to supply chain issuesImportance of direct integration of diagnostic insights with work ordersChapters00:00 Introduction to Predictive Maintenance11:46 CMMS and Predictive Maintenance32:29 Challenges in Equipment Repair47:24 Data Integration and Predictive Maintenance58:02 Integrating Advanced Maintenance Technologies

Why does AI fail in the factory? Three reasons: unstructured data, fragmented data, and no governance.At MX.0 Southeast in Greenville, SC, Zack Scriven sits down with Matt from HiveMQ and Remus from Concept Reply for a field panel on why most manufacturers aren't ready for AI — and what to build first. They cover what a Unified Namespace actually is (two definitions), where it's being misused, the Excel-to-MQTT trick, AI governance as a critical skill, and how to get started when you can't even get budget approval for a consultant.00:00 — Introductions00:37 — Why AI fails: three root causes02:16 — HiveMQ as the data backbone03:01 — No data equals no AI03:21 — The Excel problem04:24 — What is a Unified Namespace?07:40 — Where UNS is misused09:03 — Don't build a UNS without a problem statement10:35 — AI as accelerant for building foundations11:52 — AI governance as a future skill12:35 — Overcoming internal resistance14:57 — People, process, technology16:34 — Start with what you controlAbelara: https://www.abelara.com#Abelara #AbelaraAscent #AI #UNS #MQTT #HiveMQ #ManufacturingTransformation #MXO➡️ Learn more: https://abelara.com/➡️ Follow us: https://www.linkedin.com/company/abelara/Subscribe for insights on digital manufacturing, leadership, and system architecture.

🔴 LIVE from ProveIt! 2026 Whiteboard Q&A with Walker Reynolds & Dylan DuFresneWednesday morning on the ProveIt! Main Stage, Walker Reynolds returns to the whiteboard, and he is joined by Dylan DuFresne, Co-Founder & Solutions Architect at Abelara.This educational style is what the industry 4.0 community loves:Real questions.Real Unified Namespace architecture.No slides. No scripts. No fluff.✔️ Live UNS mapping discussion✔️ Deep dives into enterprise IT ↔ OT architecture✔️ Real-world implementation tradeoffs✔️ Direct audience Q&ASponsored by Fuuz — the Industrial Intelligence Platform helping Fuuz the Gap between enterprise systems and the factory floor.If you care about visibility, ownership, and building systems you actually understand… this session is for you.📍 ProveIt! Conference 2026Hyatt Regency DallasFebruary 18th, 2026Subscribe for more conversations at the intersection of manufacturing, architecture, and human-first digital transformation.#ProveIt2026 #UnifiedNamespace #Industry40 #SmartManufacturing #Abelara #FUUZ #DigitalTransformation

At ProveIt! Conference 2026, Abelara did something different. We built a fictional bottling company called Cappy Hour and re-enacted a problem manufacturers face every day: data is flowing, infrastructure is in place, but leadership still can't answer basic operational questions.What ran yesterday? What's in the tanks? How much product is going down the drain?Glenn Gardner plays the president of Cappy Hour — a multi-site beverage manufacturer under cost pressure, losing market share, and facing layoffs if waste isn't found fast. Dylan DuFresne runs the architecture workshop.Ricardo Santos from Tupinix implements the plant-floor solution. And Dan Clark from Axelon shows what happens when the internal engineering team takes ownership and builds on the foundation — using AI.This session walks through the full path from problem to payback:00:00 — Introduction 03:05 — Glenn as CEO of Cappy Hour: the five languages of manufacturing 11:10 — The cost problem: why Cappy Hour is in trouble12:30 — Abelara's $25K architecture workshop13:15 — Ricardo and Tupinix: boots on the ground with Litmus Edge15:43 — Live waste tracking: $551/hour going down the drain17:48 — Phase 1 results: $29K investment, 3-month payback, $292K annual savings19:29 — Phase 2: building a composable MES with Ignition + Postgres23:33 — Dan and Axelon: AI copilot builds a downtime screen during lunch27:35 — Glenn retells the full story from the CEO's perspective29:31 — Dylan on everything else we built for ProveIt35:00 — Audience Q&AThe philosophy behind everything we do: you can't improve what you can't see. Ownership starts with visibility.*About Abelara:* Abelara works with manufacturing leaders to bridge strategy and execution. We help teams deploy modern architectures, unify disconnected systems, and unlock real operational visibility. No jargon, no fluff—just practical solutions built to scale in the real world.➡️ Learn more: https://abelara.com/➡️ Follow us: https://www.linkedin.com/company/abelara/Subscribe for insights on digital manufacturing, leadership, and system architecture.

Recorded at Fuuz HQ in Detroit, this conversation brings together Craig Scott (Founder & CEO of Fuuz), Dave Crysler (Founder of The Crysler Club), and Zack Scriven (Abelara) for a real, operator-to-operator discussion on what actually drives manufacturing results.This is not a vendor pitch. It’s a grounded conversation about how manufacturers really scale: what works, what breaks, and where technology actually belongs.*You’ll hear:*- Dave’s “planning, people, process, technology” framework and how he applies it across industries- Craig’s story building Fuuz out of real system integrator pain: too many tools, too many skill sets, too much complexity- Why good vendors say “no” when they’re not the right fit- How to think about data, ownership, and adoption without losing the humans in the system- Quick hot takes on AI: where it helps, where it distracts, and why authenticity is coming backIf you’re a plant leader, operations leader, or IT/OT leader trying to connect shop-floor reality to enterprise outcomes, this one will hit.*About Abelara:* Abelara works with manufacturing leaders to bridge strategy and execution. We help teams deploy modern architectures, unify disconnected systems, and unlock real operational visibility. No jargon, no fluff—just practical solutions built to scale in the real world.Thank you to Craig Scott and Dave Crysler for joining this Abelara Ascent podcast.➡️ Learn More about Fuuz: https://www.fuuz.com➡️ Learn More about Crysler Club: https://thecrysler.club➡️ Learn more about Abelara: https://abelara.com/https://www.linkedin.com/company/abelara/Subscribe for insights on digital manufacturing, leadership, and system architecture.Keywords: manufacturing operations, continuous improvement, lean manufacturing, digital transformation, IT/OT, ERP integration, MES, data visibility, Unified Namespace, Detroit manufacturing, system integrator, manufacturing consulting, Fuuz, The Crysler Club, Abelara, Craig Scott, Dave Crysler, Zack Scriven. #manufacturing #operations #digitaltransformation #continuousimprovement #IIoT #ITOT #MES #ERP

The conversation covers the experience of building realistic simulators in Ignition, creating detailed tag structures, and the transition from accounting to industrial controls. It also delves into go-live events, post-implementation support, and the process of improving development processes and training interns. The discussion further explores training AI models, script development, and leveraging ChatGPT for project assistance. The conversation delves into the use of AI for script creation, fine-tuning, and debugging, as well as the process of simulator creation and manual execution. It also explores the evolution of automated code generation, simulated testing in the office, and physical testing on site. Additionally, it highlights the end user experience, including simulation in PLCs, testing and collaboration on site, and the importance of ownership and accountability. The conversation covers topics such as efficient panel conversion, simulating systems with Docker, exciting developments in simulation and automation, handling complex routing and indirection in PLCs, testing and simulation workflow, and DevOps principles and project management practices. The speakers discuss the benefits of rapid panel conversion, challenges with external integrators, automated OPC module creation, and the utilization of graph databases for complex systems. They also emphasize the importance of testing and simulating systems before commissioning and highlight the alignment of project management practices and complementary approaches to project management.TakeawaysBuilding realistic simulators in IgnitionTraining and managing teamsLeveraging AI models for project optimization AI-assisted script creation and fine-tuningAutomated code generation and simulated testingEnd user experience and ownership Efficient panel conversion processUtilizing Docker for system simulationExciting developments in simulation and automationChapters00:00 Building Realistic Simulators in Ignition06:04 Experience with Ignition and Python Scripting18:13 Improving Development Processes and Training Interns23:31 Training AI and Script Development29:16 Leveraging ChatGPT for Project Assistance45:34 Automated Code Generation54:47 End User Experience01:01:25 Closing the Gap with Axelon01:13:42 Testing and Simulation Workflow01:23:32 DevOps and Project Management

The conversation explores the importance of historical data with context, the challenges of data collection and storage, and the role of unified namespace in data management. It delves into the need for context in time series data, the impact of historical data on decision-making, and the challenges of implementing a unified namespace. The discussion also covers the significance of understanding data consumers and the cultural and technical aspects of implementing a unified namespace. The conversation delves into the practical application of UNS (Unified Namespace) and its role in data storage, historization, and AI enablement. It explores the use of MQTT brokers, OPC servers, and time-based storage solutions, emphasizing the importance of context, historical data, and statistical analysis in manufacturing operations.TakeawaysHistorical data with context is crucial for decision-making and process improvement.The implementation of a unified namespace requires consideration of both technical and cultural aspects. UNS as a broader conceptImportance of context and historical dataPractical application of AI in manufacturingChapters00:00 The Role of Unified Namespace in Data Management36:26 Data Storage and Historization42:23 Requirements for UNS and AI Enablement

You spent half a million dollars on a unified namespace. Six months later, your plant manager can't answer a simple question: what ran on Line B yesterday?Despite the sensors. Despite the dashboards. Despite the check you wrote.In this episode, we walk through the Capybara Inc. case study — a fictional multi-site beverage bottling operation built entirely from real problems. This is the story every industrial transformation leader recognizes. The pilot that proved connectivity but delivered nothing. The integrator who charged a change order every time you asked a question about your own data. The database you paid for that you don't have the password to.We trace exactly what went wrong — and how Abelara and implementation partner Tupinix fixed it.What you'll learn:- Why a unified namespace without operational context is just an expensive screensaver- The four failure modes that kill UNS pilots before they deliver value- Why the first phase of implementation wasn't technical — it was change management- Three use cases that generated real ROI: plant floor visibility, in-process inventory, and warehouse accountability- How contextualizing existing sensor data eliminated 15% of unexplained downtime with a $200 fix- Why a 6ml overfill on a 500ml bottle was costing $4,000 per week in lost product- The open-source philosophy that separates real implementation partners from vendors holding you hostageThe company is fictional. Every single problem in this story is real.Presented at the Prove It conference.---Subscribe on YouTube: https://www.youtube.com/@AbelaraAscentWatch the full series: https://www.youtube.com/playlist?list=PLc0mbSF_NFx_rQO4rVO8kWhirbjyZHx4rListen on Spotify: https://open.spotify.com/show/5yaazbKgmavQpiYMrrCQ0JListen on Apple Podcasts: https://podcasts.apple.com/us/podcast/abelara-ascent/id1893477567

You have an MQTT broker. You have dashboards. You have data flowing from PLCs. And you still can't make any decisions. That is pilot purgatory. And it is more common than anyone in the industry wants to admit.In this episode, Ricardo Santos, Director at Tupinix, joins Zack Scriven and Dylan DuFresne to walk through what it actually takes to get value from a Unified Namespace in a CPG and food and beverage environment. Using Enterprise B, Capy Hour Inc., as the virtual factory for Abelara's ProveIt 2026 presentation, the team covers real use cases with real numbers.Chapters00:00 — Welcome and Context02:34 — Pilot Purgatory and Vendor Lock04:58 — Ricardo's CPG Background09:12 — When Digital Is Not the Answer Yet19:57 — The Capy Hour Use Cases: Waste, OEE, Quality 26:14 — CIP Optimization with Data Visibility31:47 — True OEE vs. Management OEE41:36 — Why Raw Event Data Is the Foundation43:57 — Utilities and Energy Management51:00 — Tribal Knowledge and the Retiring Expert58:16 — ProveIt 2026 and What to ExpectConnect with Ricardo Santos: https://www.linkedin.com/in/ricardomarquessantos/Connect with Dylan DuFresne: https://www.linkedin.com/in/dylan-dufresne-solutions/Connect with Zack Scriven:https://www.linkedin.com/in/zackscriven/Tupinix: https://tupinix.com/Abelara: https://abelara.comSubscribe to Abelara Ascent: https://www.youtube.com/@AbelaraAscent#Abelara #AbelaraAscent #Industry40 #ManufacturingTransformation #DigitalTransformation #CPG #FoodAndBeverage #UnifiedNamespace #OEE #Bottling

The podcast features a discussion among industry experts in manufacturing, focusing on the evolution of technology, the impact of Ignition, and the collaborative nature of the industry. The conversation delves into the challenges faced by end users, the importance of community, and the need for education and guidance in adopting new technologies. The conversation delves into the challenges and opportunities of enterprise-scale digital transformation, focusing on the concept of Unified Namespace (UNS) and its impact on manufacturing and industrial operations. The discussion emphasizes the importance of intentional deployment, data management, and the role of technology in driving value and transformation. The conversation covers the strategic relationship between Fuuz and Inductive, the role of data model-driven solutions, the importance of standards, the value of AI in factory connectivity, and the responsible use of AI and UNS. It also explores the impact of MQTT and Sparkplug in transforming architectures and the significance of leveraging AI and UNS for customers.TakeawaysIndustry 4.0 and digital transformationCommunity collaboration and open standards Enterprise-scale digital transformation requires intentional deployment and a focus on driving value through technology.The concept of Unified Namespace (UNS) plays a crucial role in creating interoperability and data management in industrial operations. Strategic relationship between Fuuz and InductiveRole of data model-driven solutionsImportance of standards and AI in factory connectivityResponsible use of AI and UNSImpact of MQTT and Sparkplug in transforming architecturesChapters00:00 Introduction to the Full Stack of Manufacturing05:48 Legacy Pain Points and Modern Tools13:15 The New Paradigm in Manufacturing20:18 The Role of Fuuz in Enterprise Layer28:18 Challenges of Enterprise-Scale Deployment33:18 Unified Namespace (UNS) and Interoperability43:29 Success Stories and Adoption of UNS52:08 Mindset and Approach to Technology57:59 Value of AI and UNS for Customers01:15:14 Impact of MQTT and Sparkplug in Transforming Architectures

Summary Dylan DuFresne and Zack Scriven go tool-by-tool through the five platforms that run most serious industrial stacks — Ignition, Flow Software, Litmus Edge, HighByte, and Fuuz — and answer the question every IT/OT team is asking: when do I use which, and why. This is Episode 2 of the Prove It series on Abelara Ascent, building toward the live reference architecture at the Prove It conference.About the Speakers Dylan DuFresne leads architecture and digital transformation engagements at Abelara. Zack Scriven hosts Abelara's podcast and livestream content. Both are preparing Abelara's Prove It booth — a full Enterprise B reference architecture running locally on-site at the conference.Key TopicsThe "motor overheating" problem that explains what data ops actually isEach platform's native home turf — composable SCADA, analytics, edge, integration, cloud-native MESHighByte vs Flow: data in motion vs data at restThe Prove It Enterprise B architecture — 3 sites, 5 platforms, different combinations per siteFuuz security: single outbound tunnel, no third-party surfaceRapid-fire: when NOT to use each platformPricing reality across all five vendorsSESME / SMIPS: the interoperability initiative for shared data modelsKey TakeawaysStart with a problem, not a technology. Every digital transformation that skips this tends to fail.Data in motion → HighByte. Data at rest → Flow. Device connectivity at scale → Litmus. Plant-floor visibility → Ignition. Cloud-native enterprise backbone → Fuuz.For small companies without process maturity, a clipboard and Excel beat any of these tools. People, process, technology — in that order.Notable Quotes"HighByte is for data in motion and Flow is best for data at rest." — Dylan DuFresne"For the really small companies, a good process is way more important than the tools you're using." — Dylan DuFresne"It's people, process, and technology in that order for a reason." — Dylan DuFresneTimestamps[00:00] — Welcome + Prove It series recap[04:44] — The five platforms and their native homes[15:01] — Why more platforms doesn't mean more data models[18:35] — HighByte vs Flow: data in motion vs at rest[23:19] — The Prove It Enterprise B architecture walkthrough[28:26] — Fuuz security: the single outbound tunnel[37:24] — Rapid fire: when NOT to use each platform[44:23] — Pricing reality check[48:54] — SESME / SMIPS and the interoperability vision

Abelara Lead Architect Dylan DuFresne walks Zack Scriven through the multi-site industrial automation stack Abelara planned and deployed for its Prove It 2026 sponsorship — naming specific vendors at every layer of the stack and explaining the architectural reasoning behind each choice. Starting at the edge (Opto 22 controllers), moving through site-level infrastructure (Ignition, TimescaleDB, Flow Software, TimeBase), into data operations (HighByte, Litmus Edge), across the enterprise UNS (HiveMQ broker), and up through multi-site enterprise applications (Fuuz, MaintainX, Google Cloud), Dylan builds the architecture layer by layer and takes live audience questions throughout. The through line isn't the specific tool list — it's the *why* behind each decision. Why a separate tag server in Ignition 8.3 for DevOps isolation. Why HighByte for data in motion but Flow Software for data at rest. Why vision for SCADA but perspective for MES. Why the blue-vs-red namespace distinction matters. Why Fuuz sits where it sits. The result is one of the clearest end-to-end walkthroughs of a real, deployable Industry 4.0 stack you'll find. Three things to take away 1. "Edge" means whatever the vendor wants it to mean — know what they mean.** Dylan's working definition: *"Edge is shorthand for as far into the stack as I care to look."* To a cloud vendor, the entire plant is the edge. To an OT engineer, it's the PLC. When a product labels itself "edge," you have to ask which edge. 2. HighByte for data in motion, Flow Software for data at rest.** The two tools live at the same layer of the stack, but do different jobs. HighByte is the ETL / data ops engine — pulling data from disparate sources, modeling it, moving it. Flow Software is the analytics engine — pulling data at rest from multiple stores and calculating KPIs against it. Not an "or" decision; an "and" decision. 3. Fuuz is to the enterprise what Ignition is to the site.** If Ignition is the Swiss Army knife platform for the plant floor — SCADA, MES, IoT connectivity — Fuuz is the same kind of platform for L3/L4 enterprise apps: iPaaS foundation, MES, WMS, supply chain. Different jobs, different data models, both Swiss Army knives at their respective layers. The stack, in one pass - Device layer (L0/L1): Opto 22 controllers - Edge connectivity: Ignition Edge (IOT licenses), running on bare-metal Linux or Docker - Site SCADA: Ignition Vision (for local multi-monitor operator screens) - Site MES: Ignition Perspective (for browser-based MES frontends) - Site tag server: Ignition 8.3 dedicated tag server — single source of truth for site current state - Historian: TimeBase (Flow Software) — telemetry, scalar data, in-flight time series - MES database: TimescaleDB on Postgres — contextual event data, state changes, tabular MES functions - Analytics: Flow Software — connects all the above, calculates KPIs, contextualizes - Data ops (Site A): HighByte — ETL from tag server to UNS broker - Data ops (Site B): Litmus Edge — standalone data pipeline, no local SCADA/MES needed - Enterprise UNS broker: HiveMQ (the Prove It shared broker) - CMMS: MaintainX - Enterprise apps / L3: Fuuz — MES, WMS, supply chain, ERP integration, multi-site consistency - Data warehouse: Google BigQuery / Google Cloud - Deployment: Portainer Timestamps:00:00 — Intro + the planned stack05:11 — The credibility moment: "All 8 title sponsors, by coincidence"07:45 — Edge devices, Opto 22, and why IPCs run Linux13:00 — "Edge is shorthand for as far into the stack as I care to look"15:16 — The Ignition 8.3 tag server18:02 — TimeBase vs. TimescaleDB, MES core vs. MES custom24:51 — SCADA in Vision, MES in Perspective — why27:31 — Medallion architecture, HighByte, and the enterprise UNS34:34 — Fuuz as the enterprise Swiss Army knife45:09 — Q&A

Summary Zack Scriven sits down for an impromptu conversation with Dylan DuFresne (Lead Architect & Co-founder, Abelara) on three of the industry's most-abused terms: "digital transformation," "unified namespace," and "Industry 4.0." Dylan's core argument is that the terms themselves are buzzwords — vessels that vendors fill with whatever they're trying to sell. The only useful question is what problem are you actually solving, and the only useful follow-up is what is this vendor actually selling me.The conversation moves from meta-critique into practical territory in the second half: Dylan walks through what a UNS has actually meant for the industry over the last decade (short answer: education), whether greenfield and brownfield sites need different answers (yes), and which tools he'd pick today if he were starting from scratch. It closes with one of the cleanest vendor-selection frameworks you'll hear on a podcast — and a discovery-call CTA for anyone who wants to talk through their own architecture.Three takeaways "Digital transformation" is whatever the vendor needs it to be to close the sale. The term has meant ERP-to-cloud migration, workflow digitization, knowledge-base digitization, and now unified namespace. The throughline isn't technical — it's commercial. If you're buying "digital transformation," demand to know what product or service is actually being delivered, and whether you have a problem it solves.UNS is a tool, not an answer. The real value of the UNS movement was education. Over ~10 years, the UNS conversation did something no one argued against: it educated the market on the need to connect ERP to plant floor, and share data across layers that never shared it. But "I want a UNS" is still the wrong starting question. Greenfield deserves a UNS-shaped architecture. Brownfield with working historians usually does not.Pick tools by problem, pick vendors by partnership. Dylan's ideal greenfield stack: Ignition, HiveMQ, Flow Software, Solace — some combination gets you most of what you need. Broker market has commoditized; the differentiation is adjacent features and culture. When two vendors solve the same problem, pick the one who wants to grow with you — not the one with the nicer salesperson.Chapters00:00 — Cold open + intro 00:32 — What "digital transformation" actually means (and who benefits from the ambiguity) 03:56 — Why digital transformation is change management, not a product 05:08 — Unified namespace: what it is, what it became, what it's worth 08:30 — Greenfield vs. brownfield — and the real value of UNS 14:56 — Who actually benefited from the UNS movement 17:27 — The stack Dylan would build today (+ CoreFlux deep dive) 19:24 — How to pick vendors when products overlap 24:41 — Discovery calls + weekly architecture Q&A

SummaryGlenn Gardner (Abelara co-founder) sat down with Craig Scott (Fuuz CEO) at Fuuz HQ after spending a week embedded in bootcamp with the Fuuz team. The conversation threads through three topics: the unusual culture Fuuz has built around genuinely caring about manufacturing, the architectural decision to be the MES "shim" between ERP and the plant floor, and how democratized plant-floor data changes work for the personas nobody usually builds for — finance, accounting, and maintenance.About the guestCraig Scott — CEO and founder of Fuuz, a manufacturing operations platform purpose-built to unify ERP, plant floor, and operational workflows in a single data layer. Background spans CNC and robot programming, owning a manufacturing shop, and running a systems integration practice before Fuuz.Three takeawaysCulture is a hiring choice, not a slogan. Fuuz didn't hire developers who already knew manufacturing. They hired strong developers and put them on real implementation projects, forcing them to learn manufacturing by solving actual customer problems. Their CTO Lance now runs a maker space on the side.ERP and plant floor are different data models — stop trying to merge them. ERP is relational, transactional, 300 transactions/day. Plant floor is scalar time series, thousands of data points per second. Predictive maintenance is vector matrix tensor. You can't force all three through one pipe. What's needed is a unified data layer (the "shim") that speaks all three natively.The forgotten personas are the highest-leverage users. Finance and accounting teams burn cognitive cycles trying to reconstruct variances with spreadsheets. Maintenance teams can't get their predictive data to the enterprise. A platform that democratizes plant-floor data to those personas is worth more than another BI dashboard.Notable quotes"I found the park of people like me." — Glenn Gardner, on the Fuuz team's culture"ERP doesn't have to be those things, but there has to be that shim, that MES layer in there." — Craig Scott"I still have ulcers from those periods." — Glenn Gardner, on quarter-close variance huntingChapters00:00 — Cold open: MES theory vs. reality00:38 — The Bumblebee culture metaphor03:37 — Why Fuuz hired devs and taught them manufacturing09:26 — What kind of SI Fuuz was, and how Fuuz emerged13:00 — The MES shim explained (ISA-95 / Purdue layers)17:11 — Why iPaaS approaches fail19:10 — Scalar vs. time-series vs. tensor data21:57 — 300 ERP transactions vs. millions of telemetry events24:08 — The AWS outage and Fuuz's architectural resilience25:11 — Predictive maintenance and the vector-tensor database problem28:25 — Democratizing data to finance and accounting31:19 — The "ulcers" story — quarter-close variance chasingLinks & resourcesFuuz: https://www.fuuz.comAbelara: https://www.abelara.comCraig Scott on LinkedIn: https://www.linkedin.com/in/craigascott1/

SummaryThe Abelara team — Glenn Gardner, Zack Scriven, and Dylan DuFresne — just spent a week at Fuuz corporate headquarters for bootcamp. In this roundtable debrief, they give their unfiltered take: what they loved, what needs work, how Fuuz compares to Ignition and Plex, where it fits in the stack, and why a greenfield site should probably run both Ignition and Fuuz together. The conversation covers real pricing comparisons, the MCP tooling announcement, and why "it does everything" is both Fuuz's greatest strength and its biggest go-to-market challenge.Key Topics CoveredWhat each team member is most excited about after bootcampWhat Fuuz needs to improve: UX, user journeys, discoverability of featuresFuuz vs. Ignition: not competitors — different layers of the stackFuuz vs. Plex: why enterprises evaluating Plex should look at FuuzPricing reality: single-site MES is comparable to Ignition; multi-site is where Fuuz pulls aheadEdge gateway architecture: cloud platform with an on-prem edge componentMCP tooling: Fuuz is building it now while others have only announced itThe three pillars: schema designer, flow designer, screen designerWhy "it does everything" is both the value prop and the sales problemBuild vs. buy spectrum: Fuuz as the middle ground between full custom and rigid off-the-shelfKey TakeawaysIgnition owns the lower stack (SCADA, plant floor). Fuuz owns the upper stack (enterprise MES, iPaaS, ERP integration). A greenfield site should run both — they're best friends, not competitors.Fuuz's hardest problem isn't the product — it's explaining the product. It's six or seven commercializable products in one platform, and nobody can pitch that in a sentence without sounding like BS.MCP tooling on the Fuuz flow designer means a lotNotable Quotes"Ignition and Fuuz should probably be the best of friends." — Dylan DuFresne"It's like seven different applications. And if someone said pitch me on Fuuz, I would sound like an idiot." — Glenn Gardner"Other leading platforms have announced the ability to do this in the future. Fuuz is doing it now." — Dylan DuFresne (on MCP)Timestamps / Chapters[00:00] — What stood out most from Fuuz bootcamp [01:56] — Zack: enterprise orchestration and best-of-breed integration [03:08] — Dylan: the people, the platform, and where Fuuz fits in the stack [04:48] — Honest critique: what needs to improve [06:48] — The discoverability problem and the change logging trap [09:31] — Edge gateway explained: cloud platform with edge component [13:02] — "Fuuz is like six or seven different products" [15:12] — Pricing reality: Fuuz vs. Ignition vs. enterprise iPaaS [17:47] — Single-site vs. multi-site: where Fuuz takes the lead [20:06] — Ignition and Fuuz: complementary, not competitive [21:44] — Fuuz vs. Plex: entirely different platforms [23:29] — What module would you add next? SCP. [24:40] — The most common problem: integrating ERP with the plant floor [32:55] — MCP tooling: the AI unlock [33:41] — Glenn's final take: card-carrying Fuuz fanLinks & ResourcesFuuz: fuuz.comAbelara: abelara.comIgnition: inductiveautomation.com

SummaryGlenn Gardner frames the quality problem every manufacturer faces: DPPM spikes and nobody can figure out whether it's a catastrophe or a blip — because the data lives in a dozen different systems. Steve Modrall walks through how Fuuz solved this for HighBar, the world's first solar-powered steel mill, deploying MES, WMS, yard management, scale house, and vendor portal in nine months. Craig Scott demos Fuuz's schema designer and shows how embedded quality checks and digital work instructions eliminate the "more training" corrective action that plagues every 8D process.Key Topics CoveredQuality as the hardest problem in manufacturing: tracking DPPM across downstream, midstream, and upstreamWhy ERP falls down for midstream and upstream quality dataHighBar Steel: world's first solar-powered steel mill, 2x output per FTE vs. industryRed data vs. blue data: enterprise governance vs. plant-level detailiPaaS as the bridge between ERP and plant floor (not point-to-point)WIP visibility without SKU proliferationSchema designer: modeling manufacturing data as normalized relationshipsPoke-yoke through MES: embedding quality into process, not as an afterthought8D root cause failures when you run out of dataMCP integration for AI/LLM analysis on manufacturing dataKey TakeawaysThe ERP handles downstream data (RMAs, customer returns) well, but for midstream (factory floor) and upstream (supplier quality), you need MES, SCADA, and quality systems — the ERP was never designed for that level of detail.Don't track WIP in the ERP — it explodes your SKU count by an order of magnitude. Keep raw materials and finished goods in ERP; let the MES own the WIP detail.If your 8D corrective action says "more training," you've run out of data. Embed quality checks into the process so the failure literally cannot propagate forward.Notable Quotes"Your business doesn't happen in the ERP. It happens on the plant floor." — Zack Scriven (quoting Walker Reynolds)"You start making stuff up because you're out of data." — Glenn Gardner (on 8D failures)"They're running this mill with half the staff and generating double the output per FTE." — Craig Scott (on HighBar)Timestamps / Chapters[00:16] — Welcome and introductions [01:34] — Glenn: the manufacturing reality check [03:02] — Quality as the most direct impact to revenue, margin, and customer sat [05:14] — Downstream data: ERP handles RMAs well [06:30] — Midstream: where ERP falls down [08:20] — Upstream: supplier quality[12:08] — "Your business happens on the plant floor" [13:22] — Steve: HighBar Steel case study [15:35] — Fuuz iPaaS: agnostic integration vs. point-to-point [24:08] — Craig: MCP and AI/LLM on manufacturing data [27:47] — Why serial-level detail doesn't belong in the ERP [30:28] — Fuuz as the unified namespace for this architecture [34:34] — WIP in ERP = SKU hell [36:21] — Red data vs. blue data explained [40:45] — Craig: live demo of the schema designer [54:00] — Poke-yoke through MES [57:09] — ISO 9001: embedded procedures vs. printed SOPs [59:29] — The drunk guy looking for his keys: why 8Ds failLinks & ResourcesFuuz: fuuz.comAbelara: abelara.comOracle NetSuite: netsuite.com

SummaryGlenn from Abelara visits Fuuz corporate headquarters during boot camp week and sits down with Steve to walk through the real manufacturing pain points that Fuse solves — from the supply chain and factory floor divide, to quality traceability, bill of materials wars, real-time variance tracking, and the multi-plant standardization trap. Steve frames Fuse as the system that lives "where the carpet meets the concrete."About the GuestSteve is a senior leader at Fuuz with deep manufacturing operations experience. He positions Fuse as a composable platform covering MES, WMS, quality, scheduling, EDI, and iPaaS integration — designed to complement ERP rather than replace it.Key Topics CoveredThe supply chain vs. factory floor divide: pegging sales orders to work ordersERP's visibility gap below the shop floor levelQuality traceability: tracking components from dock to shipmentRecording PLC parameters per serial number for root causeThe bill of materials war: supply chain, manufacturing, and R&D viewsThe golden triangle: PLM ↔ MES ↔ ERP integrationReal-time variance tracking by work center and operatorMulti-plant standardization without forcing ERP onto the floorReducing tech stack from 40 vendors with a composable platformBalancing corporate control vs. local plant innovationKey TakeawaysPegging sales orders to work orders lets planners make partial shipment decisions in real time — most systems produce blanket work orders blind to customer demand.The million-unit recall problem exists because companies test everywhere but trace nowhere — traceability is the missing layer, not more testing.Use ERP for financials and front office; use MES for the shop floor. Forcing ERP down to the plant floor is why no enterprise has ever achieved one-ERP standardization.Notable Quotes"Where the carpet meets the concrete — that's where Fuse takes over." — Steve"The money's made and lost on the shop floor." — Steve"No two machines are alike. Even from the same manufacturer, they're all different." — SteveTimestamps / Chapters[00:00] — Inside Fuse HQ: a software company that knows manufacturing[01:34] — The supply chain and factory floor divide[03:14] — Pegging sales orders to work orders[04:30] — ERP's visibility gap on the shop floor[06:55] — Quality: the hardest challenge in manufacturing [08:09] — Tracking components from dock to shipment [09:48] — The million-unit recall problem[11:05] — Recording 120 PLC parameters per serial number [12:48] — What R&D leaders really need from production data [14:18] — The bill of materials war[15:29] — The golden triangle: PLM, MES, and ERP[18:53] — Real-time gross margin and variance tracking [21:42] — IT's challenge: future-proofing manufacturing [24:04] — The multi-plant standardization trap[27:49] — Employee development across standardized plants[29:48] — Closing thoughtsLinks & ResourcesFuse: fuse.comSteve Modrall: LinkedInGlenn Gardner: LinkedInAbelara: abelara.com

Dylan DuFresne walks through Abelara’s reference architecture for modern industrial systems — from edge controllers through Ignition’s tag gateway pattern, UNS publishing, historian integration with Timebase, Flow Software for contextualization, and enterprise-scale deployment with HiveMQ cloud bridging and Fuuz as an MES layer. The episode covers how to phase these components in with zero technical debt and how Abelara helps enterprises cut through analysis paralysis.About the GuestDylan DuFresne is an architect at Abelara specializing in Ignition, MQTT / Unified Namespace, and enterprise data systems.Key Topics CoveredTag gateway pattern: separating tag providers from front-end / back-end DevOpsIgnition 8.3 improvements and Gateway Area Network updatesHistorian integration: Timebase direct from tag gateway vs collectorsFlow Software for contextualization, KPIs, and bidirectional flowBlue namespace vs red namespace: site flexibility vs enterprise standardsHiveMQ cloud bridging to Snowflake, BigQuery, AzureFuuz as an MES layer replacing ETL between ERP and site systemsPhased implementation: SCADA → UNS → historian → analytics → AIKey TakeawaysUse a dedicated tag gateway so front-end and back-end systems can run DevOps cycles without risking production tag state.Standards should get stricter as you move up the stack — fluid near controllers, rigid near ERP.Build additively with zero technical debt: SCADA, then UNS, historian, analytics, AI.Timestamps[00:00] Tag gateway pattern[02:42] Timebase historian[04:01] Flow Software[07:45] HiveMQ + cloud data lakes[13:04] Fuuz as enterprise MES[15:06] Zero technical debt roadmap[17:50] Abelara workshopsLinksAbelara: https://abelara.comIgnition: https://inductiveautomation.comFlow Software: https://flowsoftware.comTimebase: https://timebased.ioHiveMQ: https://hivemq.comFuuz: https://fuuz.comLitmus: https://litmus.ioAxilon: https://axilon.com

Dan Prudhoe spent 15 years building industrial systems at a large chemical manufacturer — scaling from beta Ignition in 2009 to 70+ gateways across 30+ sites. In this ICC 2025 conversation, he walks through replacing a legacy Unix MES with Ignition, adopting Spark Plug B and MQTT, hitting MQTT's limitations, and building an enterprise event mesh with Solace that streams millions of tags from edge to cloud.About the GuestDan Prudhoe is a Senior Solutions Engineer at Solace. Previously he spent 15 years at a large chemical manufacturer leading Ignition adoption from beta, building a standardized platform across 30+ sites, and architecting an event mesh for millions of historized tags.Key Topics CoveredIgnition from beta (2009) to enterprise-scale across 30+ plantsBuilding modular, upgradeable standards across diverse sitesSpark Plug B and MQTT: decoupling systems, lightweight data collectionMQTT limitations for transactional and guaranteed deliveryEvent mesh with Solace: edge to cloud, multi-protocol, guaranteed messagingSelf-service analytics with SeeqUNS debate: centralized semantic layer vs. composable hierarchiesGraph databases and industrial data modelingThe path: self-service → automated → autonomous (agentic AI)Key TakeawaysStart with analytics, not SCADA — lower risk, proves value fast, builds political capital for mission-critical systems later.Build standards designed to flex — the first few deployments will break your standard, and that's expected.The path to autonomous manufacturing: self-service (SMEs solve their own problems), automated (event-driven workflows), then autonomous (agentic AI investigates and prescribes).Notable Quotes"We wanted to build a standard that's meant to have flexible standards." — Dan Prudhoe"Instead of we don't have enough data, they're almost like oh my god there's so much data." — Dan Prudhoe"We were able to allow them to start solving their own problems." — Dan PrudhoeTimestamps / Chapters[00:00] — Who is Dan Prudhoe? [01:06] — Ignition beta testing in 2009 [02:20] — MES 101 [05:10] — Scaling to 70+ gateways across 30 sites [08:53] — In-house vs. system integrator [11:01] — Ignition becoming enterprise-grade [12:00] — From Spark Plug B to event mesh [15:14] — MQTT's limitations [16:51] — Solace: edge to cloud architecture [20:55] — Self-service analytics for SMEs [26:31] — Predictive maintenance data provisioning [28:30] — The UNS debate [32:08] — Graph databases and ontology [33:42] — Wrap-upLinks & ResourcesSolace: solace.comDan Prudhoe LinkedInIgnition: inductiveautomation.comFlow Software: www.flow-software.comSpark Plug B: sparkplug.eclipse.orgAzure Data Explorer: azure.microsoft.com/en-us/products/data-explorer