A legacy MES blocks AI because it holds production data in closed, heavily customized databases with no modern APIs, so models can’t get clean, real-time data in or send decisions back. Most AI use cases in a plant depend on exactly that data.

• The market is moving to cloud, composable MES. MarketsandMarkets projects the MES market will grow from $15.95 billion in 2025 to $25.78 billion in 2030, driven by cloud MES and AI-driven analytics. Gartner’s 2025 Market Guide for MES, as reported by Tulip, names composability, low-code extensibility and interoperability as the key trends. Legacy systems are falling behind on all three.
• Customizations have frozen the system. Years of custom screens, scripts and integrations mean even small upgrades are risky. Many plants stay on unsupported versions because nobody wants to touch the code that keeps the line running.
• Shop-floor data is trapped and inconsistent. Each plant often runs its own MES version, data model and naming. Without a common structure, an AI model built for one line or site rarely works at the next.
• AI ambitions are running ahead of data readiness. In a Manufacturing Leadership Council survey cited by Deloitte, only 6% of manufacturers were using agentic AI in 2025, with about 24% expecting to within two years. Deloitte points to siloed data across new and legacy IT and OT systems, including MES, as a core barrier.
• Replacing MES outright is a big risk. A full rip-and-replace can take years and put production at risk at every go-live. OptiSol takes a different route: it modernizes legacy MES in phases with its iBEAM accelerator, so business logic is kept and the line keeps running.

An AI-ready MES gives every plant real-time, consistent production data and the ability to act on AI insights directly on the line. The move stays low-risk when it’s phased, tested in parallel and governed from the start.

• Real-time visibility across lines and plants. With a common data model, operations leaders can compare OEE, scrap, downtime and throughput across sites on the same basis, instead of reconciling different reports by hand.
• Predictive quality and maintenance. Models trained on unified process, machine and quality data can flag drift before it turns into scrap or downtime. Edge vision can feed defect data back to the MES in real time. OptiSol’s Scanflow supports this kind of line-speed quality inspection.
• Faster rollouts and upgrades. Modular, API-enabled MES services can be updated or extended without retesting the whole system. New lines and plants can be onboarded from a standard template rather than another round of customization.
• Lower risk through parallel running and clear rollback. Run modernized modules alongside the legacy MES on one line, compare results, and keep a rollback plan for every cutover. Schedule changes around planned shutdowns, and validate traceability and compliance records before switching.
• Governance and security built in. Treat OT security, access control and audit trails as part of the design, not an afterthought. OptiSol’s embedded engineers work within your team, following your plant standards and change control, with overlapping hours and local client partners in the US and UK.

You can modernize the MES you have, move to a new MES platform, or add a data layer that connects legacy systems to AI. Here’s how five options compare.

• OptiSol: Modernizes your current MES in phases with the iBEAM accelerator, using reverse engineering, API enablement and data unification, so there’s no big-bang rewrite. On top, it builds governed AI agents with elsai and uses Scanflow edge vision for line-speed quality. Best fit if you want to keep proven business logic while making the MES AI-ready.
• Siemens Opcenter: A full MES platform with low-code (Mendix) and cloud options, AI built into MES workflows and a digital thread from design to execution. It means replacing or upgrading to Opcenter. Best fit if you’re standardizing on the Siemens ecosystem.
• Tulip: A composable, no-code MES based in Boston and named in the 2025 Gartner Market Guide for MES. It offers Frontline Copilots and AI Composer for building apps, which can run alongside or in place of a legacy MES. Best fit if you want fast, operator-built apps for frontline teams.
• Critical Manufacturing: A manufacturing operations platform combining MES, data and analytics, part of ASMPT, with MES Copilot and Analytics Copilot in beta. It replaces the legacy MES. Best fit for medical device and industrial equipment plants.
• HighByte: An industrial DataOps and unified namespace layer that connects to legacy MES, historians and machines. It prepares contextualized data for analytics and AI rather than running AI itself. Best fit if you want a data layer to feed AI without changing the MES.

If you’re ready to standardize every plant on a new MES, a platform vendor can make sense, though it usually means a longer program. If your current MES still runs production well but blocks AI, modernizing it in phases keeps what works and opens it up faster. OptiSol focuses on that path, with embedded engineers working within your team.

This comparison is based on publicly available information about each company, as of October 2026.

The MES in most plants holds the richest production data a manufacturer has, and that data will decide how far AI on the shop floor can go. Plants that open it up line by line, keeping the logic that works and rebuilding what doesn’t, can start using AI on quality, downtime and scheduling well before a full replacement would have gone live. The question for most operations leaders is which line to start with. OptiSol helps manufacturers make that first move and the ones after it, with engineers working as part of their plant teams.