# Shopfloor Matters > Practical, vendor-neutral writing on the smart factory, shopfloor AI and the MES/MOM systems underneath them, by Matt Barber (VP & General Manager, MES at Infor; former MES developer, implementation lead and head of product). Covers MES, MOM, ISA-95, OEE, connected worker, integration with ERP, PLM, WMS, APS, QMS, LIMS and CMMS, implementation and the business case. Cite as: Matt Barber, Shopfloor Matters, https://www.shopfloormatters.com. Quoting with attribution and a link is welcome. ## Start here - [How to modernise a factory, step by step](https://www.shopfloormatters.com/post/how-to-modernise-a-factory): Modernise step by step. Assess where you really are, map the process, put a number on the problem and build the team. Then fix the master data, choose a platform for the long term and start with one site, with the end in mind. Connect the machines, go live and scale with a template. - [ERP, MES, MOM and the rest of the stack: what each one does](https://www.shopfloormatters.com/post/mes-mom-erp-compared): ERP runs the business, MES runs production on the shopfloor, and MOM runs the whole operation across production, quality, inventory and maintenance. APS, WMS, QMS, LIMS, EAM, PLM and SCADA each add depth in one area. Agree which system owns each piece of data, and add specialists only where the operation genuinely needs them. - [Shopfloor applications explained: what each one does, and why](https://www.shopfloormatters.com/post/shopfloor-applications-explained): Shopfloor applications cover production execution and OEE, in-process quality and SPC, inventory and traceability, maintenance, tooling, energy, and connected worker tools such as work instructions, handover and skills. AI builds on all of them. Match the application to the problem that hurts most, but solve it on a platform that can grow into the rest. - [Where to start with a smart factory: your first 90 days](https://www.shopfloormatters.com/post/smart-factory-where-to-start): Use the first 30 days to understand where you really are: walk every shift, assess maturity and put numbers on the biggest losses. Use the next 30 to choose the first site, map its processes and check its master data. In the last 30, build the business case on ROI and value, create a vision people buy into, secure a sponsor and build the team. - [Learning paths](https://www.shopfloormatters.com/paths/): guided routes through the articles, in order - [FAQ](https://www.shopfloormatters.com/faq/): short answers to common questions, each linked to the full article - [Role explorer](https://www.shopfloormatters.com/roles/): what MES/MOM means for each role, from CEO to operator: KPIs, challenges and the applications that matter most - [MES compared](https://www.shopfloormatters.com/compare/): straight answers to MES vs ERP, MOM, MRP, SCADA, historian, IIoT, UNS, low-code and other comparisons - [Glossary](https://www.shopfloormatters.com/glossary/): definitions of MES, MOM, OEE, ISA-95, UNS and other shopfloor terms - [About Matt Barber](https://www.shopfloormatters.com/about): background and credentials - [Podcasts, interviews and press](https://www.shopfloormatters.com/featured) - [Full text of every article](https://www.shopfloormatters.com/llms-full.txt) ## Foundations - [CMMS vs EAM vs MES: which maintenance system does what?](https://www.shopfloormatters.com/post/cmms-vs-eam-vs-mes): A CMMS helps a maintenance team plan and record work, such as work orders, preventive schedules and spare parts, usually at one site. An EAM does that and manages assets across their whole life and across sites, including cost and capital planning. ERP covers a fair amount of that too, from the asset register and depreciation to capital spend. MES/MOM knows what the equipment is doing right now, so it supplies the run hours, counts and downtime reasons that maintenance planning depends on. - [First pass yield vs rolled throughput yield: which one tells the truth?](https://www.shopfloormatters.com/post/first-pass-yield-vs-rolled-throughput-yield): First pass yield (FPY) is the share of units that pass a step right first time, with no rework or scrap. Rolled throughput yield (RTY) multiplies the FPY of every step to give the probability a unit gets through the whole process right first time. Final yield counts reworked units as good, so it can look far better than RTY. - [MTBF vs MTTR vs MTTF: maintenance metrics explained](https://www.shopfloormatters.com/post/mtbf-vs-mttr-vs-mttf): MTBF (mean time between failures) is how long a repairable asset runs, on average, before it fails. MTTR (mean time to repair) is how long it takes to get it running again. MTTF (mean time to failure) is how long a non-repairable item lasts before it's replaced. MTBF and MTTR combine into availability, and all of them depend on accurate downtime data. - [PLC vs SCADA vs DCS vs MES: the automation stack explained](https://www.shopfloormatters.com/post/plc-vs-scada-vs-dcs-vs-mes): A PLC controls a machine in milliseconds. SCADA supervises equipment, showing operators what it's doing and raising alarms. A DCS controls whole process plants as one integrated system. MES/MOM sits above them all at ISA-95 level 3, managing orders, materials, quality and genealogy. Most plants have PLCs, and use SCADA or a DCS depending on the process, with MES/MOM above the control layer. - [QMS vs LIMS vs MES: which quality system does what?](https://www.shopfloormatters.com/post/qms-vs-lims-vs-mes): A QMS runs the quality management system across the business, a LIMS runs the lab, and MES/MOM runs quality at the point of production, including event-triggered sampling, holds and batch release. Serious non-conformances and CAPA are best escalated into ERP. Most manufacturers need MOM and ERP; a LIMS is mainly for lab-heavy, regulated industries. - [Takt time vs cycle time vs lead time: what each one measures](https://www.shopfloormatters.com/post/takt-time-vs-cycle-time-vs-lead-time): Takt time is the pace you need to produce at to meet demand: available time divided by customer demand. Cycle time is how long it actually takes to complete a unit or a step. Lead time is how long the customer waits, from order to delivery, including all the waiting in between. Comparing them shows whether you can meet demand, and how much of lead time is waiting. - [ERP vs MRP vs MES: what each one does, and how they fit together](https://www.shopfloormatters.com/post/erp-vs-mrp-vs-mes): ERP runs the business on one set of data, from finance to purchasing. MRP is the planning process inside ERP that works out what to make and buy, and when, from demand, bills of materials and stock. MES/MOM runs that work on the shopfloor and records what actually happened. Most manufacturers use all three, with MES/MOM feeding actuals back so the next MRP run starts from reality. - [ISA-88 and ISA-95: how the two standards fit together](https://www.shopfloormatters.com/post/isa-88-and-isa-95): ISA-88 (S88, IEC 61512) is the batch control standard, first published in 1995. It defines physical, process, procedural and recipe models, and separates recipes from equipment. ISA-95 (IEC 62264), with Part 1 published in 2000, standardises how manufacturing operations connect to the business, defining the four areas of operations and the data ERP and MES/MOM exchange. ISA-88 lives mainly at the control level, ISA-95 mainly at level 3 and the boundary with ERP, and batch process plants usually need both. - [What is a unified namespace (UNS)? And where MES/MOM fits](https://www.shopfloormatters.com/post/unified-namespace-uns): A unified namespace (UNS) is an architecture for sharing live data: systems publish what they know to a central broker, organised by the plant hierarchy, and others subscribe. MES/MOM is an application that runs the operation, releasing orders, enforcing checks and keeping history with its context. A UNS shows the current state rather than running the operation, and it gives the sender no confirmation that data arrived, so it's a poor fit for transactions. - [ERP, MES, MOM and the rest of the stack: what each one does](https://www.shopfloormatters.com/post/mes-mom-erp-compared): ERP runs the business, MES runs production on the shopfloor, and MOM runs the whole operation across production, quality, inventory and maintenance. APS, WMS, QMS, LIMS, EAM, PLM and SCADA each add depth in one area. Agree which system owns each piece of data, and add specialists only where the operation genuinely needs them. - [Shopfloor applications explained: what each one does, and why](https://www.shopfloormatters.com/post/shopfloor-applications-explained): Shopfloor applications cover production execution and OEE, in-process quality and SPC, inventory and traceability, maintenance, tooling, energy, and connected worker tools such as work instructions, handover and skills. AI builds on all of them. Match the application to the problem that hurts most, but solve it on a platform that can grow into the rest. - [What is ERP, and why every manufacturer needs a good one](https://www.shopfloormatters.com/post/what-is-erp): ERP runs the business side of a manufacturer on one set of data: finance, sales orders, purchasing, planning, inventory and costing. A good one is worth every bit of effort. It plans and accounts for production rather than running it minute by minute, which is where specialist applications such as MES/MOM complement it. - [Lot vs batch vs serial traceability: how granular do you need to be?](https://www.shopfloormatters.com/post/lot-vs-serial-traceability): Lot traceability tracks quantities that share an identifier, serial traceability tracks every individual item, and time-based traceability traces continuous processes by time window. Lot and batch are often used interchangeably. The right level depends on the material and product, what customers and regulators require, and what a recall would cost. - [The smart factory glossary](https://www.shopfloormatters.com/post/smart-factory-glossary): Plain-English definitions of the terms you'll hear in smart factory projects, from APS and CAPA to SPC and UNS. Each entry explains what the term means in practice and links to the article that covers it in more depth. Bookmark it for the next time a term comes up that you want to pin down. - [Historian vs MES: what each one stores, and whether you need both](https://www.shopfloormatters.com/post/historian-vs-mes): A historian stores high-frequency machine signals, often every second, for engineers to trend and troubleshoot. MES/MOM records operational events with context, such as orders, lots, stops and reasons, and manages the work itself. They link by time and equipment. A historian alone can't run orders or trace material. MES/MOM also reads time-series values and adds context. For most manufacturers that removes the need for a separate historian. Only in rare cases, where engineers need very high-frequency data kept for years, do you need both. - [MES/MOM in regulated industries such as pharma and medical devices](https://www.shopfloormatters.com/post/mes-in-regulated-industries): In regulated industries you have to prove you made it correctly, often years later. Pharma relies on electronic batch records and review by exception, and medical devices on complete build records with qualified operators. Aerospace and automotive add serialised traceability, first article inspection and PPAP. Customisation multiplies validation effort. - [LIMS and MES: getting lab results to the line before the batch moves on](https://www.shopfloormatters.com/post/lims-and-mes): A good MOM handles sampling, lab results, SPC and certificates of analysis itself, with the batch context attached, which is the simpler option. A separate LIMS earns its place in heavy, regulated labs in certain industries, and then needs integration so samples carry their context and results flow straight back to hold or release the batch. ERP usually holds batch status for the business. - [EAM, CMMS and MES: who owns the maintenance work order?](https://www.shopfloormatters.com/post/eam-cmms-and-mes): EAM and CMMS manage assets and maintenance work orders, with ERP usually owning the financial side. MES/MOM knows what the asset is doing right now. They overlap on downtime and usage counters, so run day-to-day maintenance inside MES/MOM, which removes the need for a CMMS, connect an EAM where asset management needs one, and make sure everyone agrees on what stopped and why. - [Maintenance strategies: reactive, preventive, condition-based or predictive?](https://www.shopfloormatters.com/post/preventive-vs-predictive-maintenance): Reactive maintenance fixes things when they break, preventive works to a schedule, condition-based acts on measured wear, and predictive forecasts failures from data. The right answer is a mix chosen asset by asset. Run cheap, non-critical assets to failure, and save predictive for bottleneck assets with measurable warning signs and enough failure history. - [QMS and MES: closing the quality loop](https://www.shopfloormatters.com/post/qms-and-mes-quality): A good MOM does everything a standalone QMS does, including CAPA, customer and supplier complaints, sampling and SPC, with the production context a QMS never has. A separate QMS means a poor experience for operators, lost context and holds out of step, so it's almost always the wrong choice. Serious non-conformances escalate into ERP for the cost, supplier and customer sides. - [APS vs ERP scheduling: when ERP is enough, and where MES fits](https://www.shopfloormatters.com/post/aps-vs-mes-scheduling): ERP plans in weeks and months, usually assuming infinite capacity. APS sequences in days and hours against finite capacity, changeovers and constraints. MES/MOM takes the scheduled orders, releases them to the line, can do some basic scheduling, and records what actually happened. Not every manufacturer needs an APS; it helps most where sequencing is genuinely complex. - [WMS vs MES vs ERP: who owns inventory once it reaches the line?](https://www.shopfloormatters.com/post/wms-vs-mes-inventory): ERP owns inventory as a business asset and WMS owns it inside the warehouse. MES/MOM owns it on the shopfloor, including line-side stock and work in progress. For the most complex warehousing, particularly distribution centres, a WMS adds the most value; for simpler warehousing, ERP with the inventory and logistics capability of MOM can be enough. The boundary that matters is the line, so agree who requests, delivers, receives and records consumption. - [From CPQ to the shopfloor: building what the customer actually configured](https://www.shopfloormatters.com/post/cpq-to-shopfloor): CPQ produces a valid, priced configuration. ERP usually turns it into an order-specific BOM and routing, and MES/MOM receives a normal production order. Engineer-to-order adds PLM to the flow, plus partial release and one-off drawings, which is where a direct PLM to MES/MOM link for documents can help. - [EBOM, MBOM, as-built: who owns the BOM at each stage](https://www.shopfloormatters.com/post/ebom-mbom-as-built): The engineering BOM lives in PLM, the manufacturing BOM usually lives in ERP, each production order gets its own copy as the order BOM, and MES/MOM records the as-built: exactly which lots and serials were used. Most integration arguments come from not agreeing who owns which stage. - [PLM, ERP and MES: where the digital thread actually runs](https://www.shopfloormatters.com/post/plm-erp-mes-digital-thread): PLM owns what a product is, ERP owns the business of making it, and MES/MOM owns execution. In most manufacturers product data reaches the shopfloor through ERP, which is a sound design. A direct PLM to MES/MOM link makes sense for rich content that ERP isn't built to carry, such as 3D models and illustrated instructions, though a central document repository that every system reads from is usually the better option for documents. - [The smart factory tech stack, system by system](https://www.shopfloormatters.com/post/smart-factory-tech-stack): A smart factory runs on several systems working together. ERP and PLM sit at the enterprise layer, with APS where scheduling needs it. MES/MOM runs operations, with a WMS, EAM or LIMS alongside where needed, and connectivity and data platforms sit beneath and around. Knowing each one's job stops you over-buying or stretching a system too far. - [SPC on the shopfloor: catch the drift before it becomes scrap](https://www.shopfloormatters.com/post/spc-on-the-shopfloor): Inspection tells you a part is bad; SPC tells you it's about to be. Control limits come from your process and specification limits from the customer, so a process can drift out of control while still in spec. Put control charts and a few well-chosen rules in front of operators, at the line, in MES/MOM. - [OEE vs OOE vs TEEP: which measure answers which question](https://www.shopfloormatters.com/post/oee-vs-teep): OEE, OOE and TEEP all multiply availability, performance and quality, but measure availability against different time. OEE uses planned production time, OOE uses the hours the factory is open, and TEEP uses every hour in the calendar. Use OEE to improve the line, OOE to challenge planned stops, and TEEP to find hidden capacity before you invest. - [Historians, data lakes and data fabric: where shopfloor data lives](https://www.shopfloormatters.com/post/manufacturing-data-platforms): Warehouses hold structured data for reporting, lakes hold raw data at scale, and data fabric connects data where it already lives. MES/MOM reads and contextualises time-series data, so a separate historian is only needed in rare cases. AI can pull from all of these, but it needs to understand the data model, and the context that makes the data valuable is created by MES/MOM at the point of work. - [OPC UA vs MQTT: which one, and when](https://www.shopfloormatters.com/post/opc-ua-vs-mqtt): OPC UA is the standard for reading machines, with structure, meaning and security built in. MQTT is a lightweight publish-subscribe protocol that moves data efficiently through a broker to many consumers. They aren't really competitors. Both can be used in the same factory, so pick the right one for each job. - [Cloud vs on-premises MES: how they compare](https://www.shopfloormatters.com/post/cloud-vs-on-premises-mes): On-premises MES/MOM runs on servers you manage; cloud MES/MOM runs on a platform the vendor manages, usually with an edge component at each site. Most multi-site on-premises MES already runs in a central data centre that sites reach over the network. Cloud adds large-scale security, easier scaling and continuous updates. On-premises fits sites without reliable connectivity, and cases where data must stay in your own data centre, such as many aerospace and defence programmes. - [OPC UA, MQTT and UNS in plain English](https://www.shopfloormatters.com/post/opc-ua-mqtt-uns-explained): OPC UA is the standard for reading machines with structure and security built in. MQTT is a lightweight publish and subscribe protocol, and both can be used in the same factory, each where it fits. A unified namespace is a real-time view of operational data that applications have published to a broker, with real risks: no confirmation to the sender and no history. Connectivity moves data; MES/MOM gives it meaning. - [MES functionality: energy](https://www.shopfloormatters.com/post/mes-functionality-energy): Most factories can't see energy below the site meter. MES/MOM links consumption to the machine and the order that caused it, including idle running, so you can measure energy per piece and find waste. Start with your biggest consumers and bottleneck lines. In my experience the first improvements usually pay for the metering. - [IT vs OT: why the two worlds think differently, and where MES/MOM sits](https://www.shopfloormatters.com/post/it-vs-ot): IT manages information and puts confidentiality first; OT controls physical equipment and puts availability and safety first. OT assets last decades and can't be patched on a whim. MES/MOM sits on the boundary at ISA-95 level 3, so it needs both: IT owning the application and security, operations owning the processes and data standards. - [SCADA vs MES: what each one does, and do you need both?](https://www.shopfloormatters.com/post/scada-vs-mes): SCADA controls and supervises equipment in real time; MES/MOM manages the operation. If SCADA is mainly there for visibility, MES/MOM can often cover it and leave you with fewer systems to run. In heavier process industries with real control loops, keep both and agree a clear boundary between them. - [Manufacturing KPIs beyond OEE](https://www.shopfloormatters.com/post/manufacturing-kpis-beyond-oee): OEE matters, but it doesn't tell you everything. Make sure you can also measure first-pass yield, scrap rate, cycle time, changeover time, OTIF, MTBF and MTTR, calculated consistently and ideally to ISO 22400. Every one depends on accurate, granular data captured as it happens, not reconstructed at the end of the shift. - [IIoT vs MES: why IIoT doesn't replace MES](https://www.shopfloormatters.com/post/iiot-vs-mes): IIoT is about connecting machines and sensors so their data can be collected. MES/MOM puts that data to work: orders, traceability, quality checks, maintenance and guiding operators. Collecting machine data is basic for any MES, and MES/MOM often acts as the IIoT layer itself. Where a separate IIoT platform is used, it feeds MES/MOM rather than replacing it. - [MES vs ERP: what each one does, and why most manufacturers need both](https://www.shopfloormatters.com/post/mes-vs-erp): ERP plans and accounts for production: orders, materials, costs and dates, across the business. MES/MOM runs it on the shopfloor in real time, recording what actually happened by order, lot, machine and operator. They overlap on orders and inventory, but do different jobs for different people, and most manufacturers past the simplest operations need both. - [MES functionality: tooling](https://www.shopfloormatters.com/post/mes-functionality-tooling): Moulds, dies, jigs and fixtures are often the biggest single factor in quality, yet many factories manage them on a whiteboard. MES/MOM counts tool life from real production, tracks cavities and readiness, triggers maintenance by usage and records which tool made each part. Changeovers become planned activities instead of scrambles. - [Connected worker vs MES: standalone app or built into MES/MOM?](https://www.shopfloormatters.com/post/connected-worker-one-app-not-five): Connected worker tools digitise frontline work: checklists, shift handover, skills, work instructions and safety sign-offs. MES/MOM runs production and quality. Connected worker can be bought as a standalone app, which is quick to start but lacks production context, or built into MES/MOM on the same data. My preference is built in. - [MES functionality: maintenance](https://www.shopfloormatters.com/post/mes-functionality-maintenance): Maintenance belongs in the same platform as production because both teams care about the same downtime. In MES/MOM, operators raise faults from the screen they already use and see the planned work and autonomous maintenance tasks for their equipment. MOM maintenance replaces the need for a separate CMMS, while an EAM still makes sense for asset-heavy operations with advanced needs. - [Your traceability probably isn't as good as you think](https://www.shopfloormatters.com/post/manufacturing-traceability): Order-level traceability in ERP does the business-level job well, but it's nowhere near what's possible at the point of work. MES/MOM gives you genealogy forwards and backwards, down to serials and exact lot quantities, plus tooling, quality results, context and process parameters. In a recall, that's the difference between minutes and a weekend in spreadsheets. - [Discrete vs process manufacturing: the types, and why they matter for MES](https://www.shopfloormatters.com/post/discrete-vs-process-manufacturing): Discrete manufacturing makes countable items that could in theory be taken apart, and includes repetitive, job shop and assembly. Process manufacturing transforms materials that can't be separated again, in batches or continuously. Mixed-mode combines the two, for example a batch that's then filled and packed. The type shapes units of measure, traceability, KPIs and which MES/MOM fits. - [MES vs MOM: why the difference matters when you buy](https://www.shopfloormatters.com/post/mes-vs-mom): MES runs production: orders, counts, scrap, downtime and OEE. MOM covers all four areas ISA-95 defines for manufacturing operations, which are production, quality, inventory and maintenance. A MOM platform covers them on one system. MES is one part of MOM, not a synonym for it, and buying a separate system per pillar often means reconciling them, unless the depth of a dedicated system justifies it. - [MES functionality: the complete guide](https://www.shopfloormatters.com/post/mes-functionality-guide): MES/MOM covers far more than production. Its core functions are production, quality, inventory and maintenance, on one set of data, and the broadest platforms add capabilities such as tooling and connected worker. Each function depends on the others, so when they share one platform the context is captured automatically. This guide links to a detailed article on each function. - [MES functionality: warehouse management and logistics](https://www.shopfloormatters.com/post/mes-functionality-logistics): Factory logistics covers receiving goods, replenishing the line, putting finished goods away, warehouse moves and shipping. It can run in ERP, a dedicated WMS or MES/MOM. ERP suits simple warehouses, a WMS suits complex or automated ones, and MES/MOM suits flows driven by production. Many manufacturers combine them, with MES/MOM running from the supermarket to the line. - [MES functionality: inventory management on the shopfloor](https://www.shopfloormatters.com/post/mes-functionality-inventory): MES/MOM manages inventory once it reaches production: lots, serials, containers and handling units at each line-side location. It validates every item at the point of use against the order and its bill of materials, and records consumption at lot or serial level. ERP keeps the totals and the value; the shopfloor system keeps the detail. - [MES functionality: quality operations](https://www.shopfloormatters.com/post/mes-functionality-quality): MES/MOM runs quality at the point of production. It manages control plans, prompts the right inspection when it's due, captures variables and defects, often straight from gauges, applies SPC and places holds on suspect material. Every result is tied to the order and the lot, which makes traceability and root cause analysis much faster. - [MES functionality: production monitoring](https://www.shopfloormatters.com/post/mes-functionality-production): Production monitoring is the most fundamental MES function. MES/MOM runs production orders on the line and records counts, scrap, rework, downtime and labour against each one, ideally straight from the machine. That data feeds OEE and continuous improvement. A good system also prevents mistakes as they happen, rather than only recording them afterwards. - [What is OEE? How to calculate overall equipment effectiveness](https://www.shopfloormatters.com/post/what-is-oee): OEE, overall equipment effectiveness, compares the good product a line made with what it could have made in its planned time. It's calculated as availability multiplied by performance multiplied by quality. Measure it at the bottleneck, capture the data automatically so short stops are counted, and use the components to see where the losses are. - [The history of MES and digitalisation in manufacturing](https://www.shopfloormatters.com/post/history-of-mes): Manufacturing systems moved from paper and skill, through PLCs in 1969 and SCADA, to ERP for the business and separate departmental systems on the shopfloor. MES emerged in the early 1990s to bring execution together, and ISA-95 in 2000 standardised the model, leading to broader MOM platforms. AI is the next step, and it grows with the same data foundation. - [The ISA-95 model explained, and where MES fits](https://www.shopfloormatters.com/post/isa-95-explained): ISA-95 (IEC 62264) is the international standard for integrating business systems with manufacturing operations. It defines five levels, from the physical process at level 0 to ERP at level 4, with MES/MOM at level 3. It also defines what ERP and MES/MOM exchange, and the four areas of manufacturing operations: production, quality, inventory and maintenance. - [MES functionality: production scheduling](https://www.shopfloormatters.com/post/mes-functionality-scheduling): ERP creates production orders from demand and stock levels. MES/MOM receives them with their routings and bills of materials, lets schedulers fine-tune the sequence and the equipment, and releases orders to the shopfloor, where operators see a live plan. A dedicated APS is worth adding when sequencing itself becomes the hard problem. - [MESA's MES models, and what ISA taking over means](https://www.shopfloormatters.com/post/mesa-and-the-definition-of-mes): MESA International, best known for its MESA-11 model of MES functionality, ceased operation on 30 June 2026, and ISA took on its educational programmes and content the same day. The MESA-11 model set out eleven functions, far broader than production execution, and later models connected MES to the wider business. - [What is MES/MOM? A guide to MES and manufacturing operations management](https://www.shopfloormatters.com/post/what-is-mes): MES, a manufacturing execution system, runs production: releasing orders and recording output and downtime as they happen. MOM, manufacturing operations management, covers the whole operation that ISA-95 defines, from production and quality to inventory and maintenance. A MOM platform does it all on the same data. The difference decides what you buy, and whether everyone in operations works from one version of the truth. ## Smart Factory & AI - [Industry 4.0 vs Industry 5.0: what's actually different?](https://www.shopfloormatters.com/post/industry-4-0-vs-industry-5-0): Industry 4.0 describes connected, data-driven manufacturing, with machines and systems integrated and analytics used to optimise operations. Industry 5.0, a European Commission concept from 2021, builds on it and adds human-centric, resilient and sustainable goals. Both depend on accurate, contextualised shopfloor data. In my view the outcomes matter more than the Industry 4.0 or 5.0 labels. - [Closing the loop: service data back to the shopfloor and engineering](https://www.shopfloormatters.com/post/service-data-back-to-engineering): When products fail in the field, the answer is often in the as-built record. Service records the failure against the serial number, ERP links it to the installed base and warranty, and MES/MOM shows what the failed units had in common. The fix then goes to whoever owns the cause, whether that's engineering in PLM or the shopfloor. - [Agentic AI and MCP, explained for people who run factories](https://www.shopfloormatters.com/post/agentic-ai-mcp-manufacturing): Agentic AI is given an objective and acts towards it, often by using other software. MCP (Model Context Protocol) is an open standard for connecting agents to applications and data. On the shopfloor, start with agents that gather information and recommend, with a person making the call, before letting them act within tight limits. - [Digital twins run on shopfloor data. MES/MOM is where it comes from](https://www.shopfloormatters.com/post/mes-and-digital-twins): A digital twin is a virtual model kept in sync with the real factory by live data. MES/MOM doesn't do the simulation, but it supplies much of the operational reality: what's running where, material, quality, asset status and people, already in context. Without that, a twin drifts away from the factory it represents. - [Eight smart manufacturing myths that refuse to die](https://www.shopfloormatters.com/post/smart-manufacturing-myths): AI does its best work when the data foundation grows with it. Excel isn't good enough to run operations. A smart factory is more than MES/MOM, ERP and MES do different jobs, and IIoT doesn't replace MES. These and three more myths come up every week, and each one falls apart once you look at how a factory actually runs. - [AI for the frontline: copilots, shift summaries and coaching](https://www.shopfloormatters.com/post/ai-for-frontline-workers): AI for the frontline should give every operator your best operator's knowledge, not replace people. Copilots, shift summaries, real-time coaching and visual inspection support all work today. They're only useful when the AI has MES/MOM context: the order, material, recent stops, approved procedures and who's asking. - [What the research says about smart factory ROI](https://www.shopfloormatters.com/post/smart-factory-roi-research): Deloitte's 2025 survey of large US manufacturers reported average gains of 10% to 20% in production output and 10% to 15% in capacity from smart manufacturing. Meanwhile over half of small and mid-sized manufacturers still rely on manual methods. What the main studies actually show, and what they don't tell you. - [Predictive maintenance needs data you probably aren't collecting yet](https://www.shopfloormatters.com/post/predictive-maintenance-data): Predictive maintenance is the last step after reactive, preventive and condition-based maintenance. Start by recording downtime and other sensor data accurately and automatically. Models then need condition data, failure history, maintenance actions and operating context, and the operating context usually lives in MES/MOM. Begin with your bottleneck assets. - [Skills you can enforce, not just record](https://www.shopfloormatters.com/post/skills-enforcement-shopfloor): A skills matrix in a spreadsheet records qualifications; it doesn't stop the wrong person running the wrong job. Define skills against specific operations and equipment, and let expiry dates and procedure changes trigger retraining. Then check qualifications at login and order start, so the system blocks the job rather than reporting it afterwards. - [You can't hit ESG targets without shopfloor data](https://www.shopfloormatters.com/post/esg-needs-shopfloor-data): ESG reporting increasingly needs evidence, and much of it starts on the shopfloor: energy per machine and product, idle consumption, scrap, yield and water. MES/MOM already knows what was made, where and for which order. Add machine-level energy data and sustainability becomes operational efficiency with a reporting benefit. - [Shift handover is a process. Treat it like one.](https://www.shopfloormatters.com/post/digital-shift-handover): Most shift handovers are a quick chat at the door, so detail gets lost. Build the handover from what actually happened, using MES/MOM data on orders, stops, quality and maintenance, then let the supervisor add what only a person knows. Give it a fixed structure with sign-off from both supervisors, and carry open actions forward until they're closed. - [Practical GenAI for the shopfloor: four uses that save time today](https://www.shopfloormatters.com/post/practical-genai-for-the-shopfloor): GenAI already saves time in factories in four ways: answering questions about operational data in plain English, helping configure systems, acting as a knowledge assistant on every screen, and drafting shift summaries. Every one depends on the AI having access to good, contextualised data from your shopfloor systems. - [Digital work instructions that know what's running](https://www.shopfloormatters.com/post/digital-work-instructions): A PDF on a tablet is paper on a screen. Proper digital work instructions appear automatically for the job that's running, are always the latest version, run step by step and capture confirmations as you go. Inside MES/MOM they can react to what's happening, which is how new starters get it right in days rather than months. - [AI, ML and GenAI on the shopfloor: what each one actually does](https://www.shopfloormatters.com/post/ai-ml-genai-on-the-shopfloor): AI is the broad category. Machine learning predicts outcomes from your historical data, and has been in factories for years through SPC and predictive maintenance. Generative AI creates content such as text and code, and helps people more than machines. Each needs different data, so ask which kind before you spend money. - [Raw machine data is worthless. Context is what makes it useful](https://www.shopfloormatters.com/post/contextualised-manufacturing-data): A machine signal that says "stopped" tells you very little. Context adds the order, product, shift, duration and reason, and that's what turns data into decisions and AI. MES/MOM is where most of that context comes from, which is why point solutions that each see one slice lose so much of the value. - [Time to truth: how long does your factory take to know what happened?](https://www.shopfloormatters.com/post/time-to-truth): Time to truth is how long it takes a factory to know what really happened, with enough context to act on it. The seven stages run from eventually (paper and Excel) and the end of the hour or shift (ERP on the shopfloor) through after the fact (digitalised point solutions), right now (automatic data capture with MES/MOM) and before it happens (harmonious systems) to before anyone asks (agentic AI with a human in the loop) and without asking (autonomous, within agreed limits). The truth arrives in order. - [What is a smart factory? A practical definition](https://www.shopfloormatters.com/post/what-is-a-smart-factory): A smart factory is a manufacturing environment where people, processes and technology work together, supported by data, to maximise efficiency, flexibility and quality. In practice that means defined processes, planning connected to execution, real-time data at the point of work, one source of truth for decisions, and people backed to make the call. - [The biggest challenges in food and beverage manufacturing](https://www.shopfloormatters.com/post/food-and-beverage-manufacturing-challenges): Food and beverage manufacturers face strict food safety demands, waste and give-away, variable yields, labour shortages and fragile supply chains. MES/MOM helps by monitoring critical parameters, enforcing recipes and procedures digitally, tracking give-away in real time and recording every ingredient lot, so traces take minutes and recalls stay narrow. - [AI will transform your factory, if you give it the right data](https://www.shopfloormatters.com/post/ai-needs-shopfloor-data): AI can transform a factory when it has the right data and context to learn from, in the same way large language models needed the internet. AI that predicts and optimises learns from the factory's own data, so it grows with the foundation: defined processes, MES/MOM on the shopfloor, connected equipment and contextualised data built up over time. The two are best built together. Generative AI can start now on documents and know-how, as long as answers are grounded and reviewed. - [The MES/MOM market: where it's heading, and what it means for buyers](https://www.shopfloormatters.com/post/mes-market-overview): The MES/MOM market is growing steadily, and most of it is still untapped: only 8% of plants use a commercial MES, and 54% still run on paper or spreadsheets. Demand is driven by workforce change, AI ambitions, traceability and composable ERP. Legacy toolkits are giving way to two modern approaches: platforms for builders, and configurable enterprise platforms. Buyers should know which type they're choosing, and why. ## Implementation - [How to modernise a factory, step by step](https://www.shopfloormatters.com/post/how-to-modernise-a-factory): Modernise step by step. Assess where you really are, map the process, put a number on the problem and build the team. Then fix the master data, choose a platform for the long term and start with one site, with the end in mind. Connect the machines, go live and scale with a template. - [Where to start with a smart factory: your first 90 days](https://www.shopfloormatters.com/post/smart-factory-where-to-start): Use the first 30 days to understand where you really are: walk every shift, assess maturity and put numbers on the biggest losses. Use the next 30 to choose the first site, map its processes and check its master data. In the last 30, build the business case on ROI and value, create a vision people buy into, secure a sponsor and build the team. - [Connecting MES/MOM to everything else: integration patterns that last](https://www.shopfloormatters.com/post/isa95-integration-patterns): Start every integration by agreeing what object moves, which system owns it, what triggers it, how fresh it must be and whether the sender needs confirmation that it was processed. ERP is usually the hub: business transactions go through ERP, and execution detail can go direct when speed or richness demands it. Then choose the pattern. Point-to-point is often the safest for transactions, because the sender gets a response. - [Code, build, stitch or configure: four ways to run your shopfloor systems](https://www.shopfloormatters.com/post/build-vs-buy-mes): There are four ways to run shopfloor systems: code your own, build from low-code blocks, stitch point solutions together, or configure an MES/MOM platform. Each has trade-offs. A configurable platform gives the most value over time, particularly for large, multi-site manufacturers. - [Rolling out across sites: the centre of excellence and template approach](https://www.shopfloormatters.com/post/mes-multi-site-rollout): Build an enterprise template that defines standard processes, master data, KPIs, screens and integrations, then roll it out site by site, allowing local differences only where there's a real reason. A centre of excellence blending IT and OT owns the template, and later sites go live much faster. - [The biggest barriers to MES success in food manufacturing](https://www.shopfloormatters.com/post/food-manufacturing-mes-barriers): Food MES projects depend on the people as much as the technology. The barriers I see most are underestimating the people and change management, starting too big, underestimating master data such as recipes and specifications, and choosing a platform that can't grow. Involve operators early and start on one line, with the data foundation right first. - [Replacing a homegrown MES: the people problem nobody plans for](https://www.shopfloormatters.com/post/replacing-a-homegrown-mes): Replacing a homegrown MES needs as much planning for the people as for the technology. Turn the developers who built it into champions and resist the urge to replicate every quirk. Don't let the 5% the new system does differently overshadow everything it does better, and map what the old system really does before you start. - [Go-live is the starting line](https://www.shopfloormatters.com/post/mes-go-live): Go-live is the starting line. Plan hypercare with support on every shift and visible super users, and fix issues fast enough to keep people's trust. Measure what the business case promised against the baseline and celebrate the quick wins. The value comes from improving and extending the system after go-live. - [Six technology mistakes that sink smart factory projects](https://www.shopfloormatters.com/post/smart-factory-technology-mistakes): Don't wire ERP straight to machines, underestimate what it takes to build your own MES platform, buy point solutions that can't scale, over-customise, add a lightweight connected worker app that collides with MES/MOM, or mistake an IIoT platform for an MES. I've seen every one of these repeatedly, and every one is avoidable. - [Resistance to change is guaranteed. Plan for it.](https://www.shopfloormatters.com/post/manufacturing-change-management): Every smart factory project meets resistance, because change is uncomfortable. Leadership has to set the vision, back change management, build a strong team, invest in skills and give the likely detractors real responsibility. Programmes across the whole landscape tend to meet less resistance than small pilots. Involve operators in the design and explain why it makes their shift easier. Adoption deserves as much investment as implementation. - [Going paperless: start with the end in mind](https://www.shopfloormatters.com/post/paperless-manufacturing): Removing paper is a great first step if you can't do everything at once. Start with the paper that hurts most, often quality check sheets or downtime logs, one process at a time. Avoid a different cheap tool for every form: choose a platform that can grow into full MES/MOM. - [How to choose an MES: 12 questions to ask every vendor](https://www.shopfloormatters.com/post/how-to-choose-an-mes): Choose MES/MOM for the long term, not just today's problem. Ask every vendor about ERP master data, integration, industry fit, operator experience, training, scope beyond production, AI and IIoT, and growth across sites. Avoid wish-list RFPs: base requirements on your process maps and separate what you need now from later. - [Configure, don't customise](https://www.shopfloormatters.com/post/configure-dont-customise): Out of the box, configuration, low-code, integrations, extensions and customisation are six different things. Only true customisation changes the product, and it costs you again at every upgrade. Your manufacturing processes are probably less unique than they feel, so pick a mature, configurable platform and challenge every gap. - [Your ERP master data isn't ready for MES](https://www.shopfloormatters.com/post/erp-master-data-for-mes): ERP master data is usually set up for costing, and MES/MOM needs more detail. ERP often has a work centre where MES/MOM needs the individual machine in the line, and it doesn't always have semi-finished material numbers at the right level as material moves through the operations. Get it right in ERP first so every system benefits, and plan MES/MOM alongside any ERP implementation or upgrade, so you don't have to rework the master data later. - [How to build an MES business case leadership will sign](https://www.shopfloormatters.com/post/mes-business-case): Business cases usually stall because of fear of overrun and failure, not the idea itself. Put money on today's pain in quality, downtime, labour, delivery and inventory, and show realistic costs including your internal people. Spend most effort on how risk is managed, and give an honest view of the cost of doing nothing. - [How long does an MES implementation take?](https://www.shopfloormatters.com/post/how-long-does-mes-implementation-take): For a first site, my answer is six to twelve months. When I polled people on LinkedIn, twelve to eighteen months was the most common experience. It can be done in four, with an educated customer, a superb team, a configurable product and a focused scope. Later sites are much faster with a template, and several sites can be done in parallel. - [Map your processes: know your as-is before you buy anything](https://www.shopfloormatters.com/post/manufacturing-process-mapping): Digitise a broken process and you get a faster broken process. Before buying anything, agree the physical hierarchy, then map what really happens using a swimlane grid of departments against processes, with the people who do the work. Decide which version of a process to standardise before you build it into software. - [Can MES/MOM improve your bottom line? Where the ROI really comes from](https://www.shopfloormatters.com/post/mes-roi-bottom-line): MES/MOM improves the bottom line mainly through less downtime, less scrap and rework, less admin time, lower inventory, smaller recalls, better ERP costing and lower energy use. Start by asking what 1% of OEE is worth on your bottleneck, measure your own losses, count costs honestly, and make it one decision for the whole programme with the delivery phased. - [Kaizen and MES: using data to drive continuous improvement](https://www.shopfloormatters.com/post/kaizen-and-mes): Kaizen is continuous improvement through small, regular changes made by everyone, with a stable standard to measure against. MES/MOM supports it by capturing losses automatically, and by showing whether an improvement has held. It won't create a Kaizen culture on its own, but it removes much of the data collection that slows improvement teams down. - [How MES takes complexity out of your ERP](https://www.shopfloormatters.com/post/mes-simplifies-erp): ERP is built to run the business, not the shopfloor. Forcing it to handle real-time execution adds custom code and complexity. Putting execution into MES/MOM alongside ERP lets both run closer to standard and improves ERP data, because MES/MOM posts actual consumption and output rather than backflushed estimates. - [Building the right team for an MES implementation](https://www.shopfloormatters.com/post/mes-implementation-team): MES/MOM projects succeed or fail on the people as much as the software. You need an executive sponsor who removes barriers, a capable project manager, IT and operations involved from the start, and operators shaping the design. Work closely with your vendor or systems integrator throughout, and plan for change from day one. - [Who benefits from MES? Every role, from operator to CEO](https://www.shopfloormatters.com/post/who-benefits-from-mes): Almost everyone in manufacturing operations benefits from MES/MOM, in different ways. Operators get guidance and less paperwork, specialists get data to plan and improve, managers get live performance and escalations, and executives get trusted figures for cost and delivery. The breadth of benefit depends on how much of the operation the product covers. ## Tools and templates - [What is 1% of OEE worth?](https://www.shopfloormatters.com/tools/oee-one-percent): Two numbers from your annual report. See what a 1% improvement in OEE is worth to your business, and what 5%, 10% or 25% would mean. - [Map your systems landscape](https://www.shopfloormatters.com/tools/systems-landscape): Pick the systems you run and how they connect. See your landscape drawn live, with the overlaps and gaps worth a conversation. ## By industry and role - [The smart factory in food and beverage](https://www.shopfloormatters.com/for/food-and-beverage): Thin margins, short shelf lives, allergens, and a recall that can arrive at any hour. Where MES/MOM earns its keep in food and drink, and the traps I see most often. - [The smart factory in automotive manufacturing](https://www.shopfloormatters.com/for/automotive): Just-in-time schedules and zero-defect expectations, from customers who can audit you. Where MES/MOM matters most for automotive suppliers. - [The smart factory in packaging manufacturing](https://www.shopfloormatters.com/for/packaging): Fast lines, frequent changeovers, short runs and customers who change the artwork on a Friday. Where MES/MOM pays back in packaging. - [The smart factory in metal fabrication](https://www.shopfloormatters.com/for/metal-fabrication): High-mix, low-volume work, and jobs that wander through the shop. Where MES/MOM helps fabricators and machine shops see what's really happening. - [The smart factory in plastics manufacturing](https://www.shopfloormatters.com/for/plastics): Moulds, cavities, cycle times and process windows. Where MES/MOM pays back in injection moulding and extrusion. - [For plant and operations managers](https://www.shopfloormatters.com/for/plant-managers): You own the numbers and the people. Where to start, how to build a case the board will sign, and how to avoid paying twice. - [For IT and digital leaders](https://www.shopfloormatters.com/for/it-leaders): You know ERP inside out. The shopfloor speaks a different language. Architecture and data, without the turf war. - [For quality managers](https://www.shopfloormatters.com/for/quality-managers): From checks on clipboards to quality built into the process. Where MES/MOM, QMS, LIMS and ERP each fit, and what to fix first. ## Learning paths - [MES/MOM foundations](https://www.shopfloormatters.com/paths/foundations): What MES and MOM actually do, how ISA-95 fits the systems together, how OEE is measured, and how MES/MOM works alongside ERP. - [MES functionality in depth](https://www.shopfloormatters.com/paths/mes-functionality): One function at a time, from scheduling and quality to inventory, maintenance and energy, with what each looks like when it runs in MES/MOM. - [Metrics that matter](https://www.shopfloormatters.com/paths/metrics): How the common shopfloor measures are calculated, and how to use them to improve performance rather than just report it. - [Quality and traceability](https://www.shopfloormatters.com/paths/quality-traceability): Traceability that stands up to a recall, quality systems compared, SPC on the line and what regulated industries need on top. - [Maintenance and reliability](https://www.shopfloormatters.com/paths/maintenance): Maintenance strategies compared, the systems that support them, the reliability metrics behind them and the data predictive maintenance depends on. - [Connected worker](https://www.shopfloormatters.com/paths/connected-worker): Giving the people on the line the right instructions, skills checks and handovers, on one screen. - [Beyond ERP](https://www.shopfloormatters.com/paths/beyond-erp): ERP runs the business. These are the strategic systems that run the operation around it, from MES/MOM and connected worker to WMS, APS and CPQ, and how to decide what you need. - [The systems landscape](https://www.shopfloormatters.com/paths/systems-landscape): How the systems actually connect. Integration patterns, who owns the BOM, quality, lab and maintenance data, and the protocols that join the shopfloor up. - [Connecting the plant](https://www.shopfloormatters.com/paths/connecting-the-plant): From PLCs and SCADA up to MES/MOM, the protocols in between, and where the data is stored. - [Shopfloor AI](https://www.shopfloormatters.com/paths/shopfloor-ai): What AI needs from your shopfloor data, where it already pays off, and what agentic AI changes next. - [Smart factory strategy](https://www.shopfloormatters.com/paths/strategy): What a smart factory really is, the myths to ignore, where to start and how to modernise without betting the plant. - [The business case](https://www.shopfloormatters.com/paths/business-case): Who benefits, where the money comes from, how to build a case leadership will approve, and how to choose without regret. - [Implementation playbook](https://www.shopfloormatters.com/paths/implementation): Every stage of a shopfloor project in the order you will hit it, from mapping the process to rolling out across sites.