Questions plant teams ask

The questions manufacturers ask when they plan a plant-wide data foundation, each answered with what MaestroHub does today, the steps to see it, and what it does not do yet.

Questions

38, in eight themes

Applies to

3.0

Each answer

Short answer, detail, try it, limits

Each question links to its full answer. The status says how much of the ask MaestroHub covers today:

• Available the whole ask is in the release.
• Beta it rests on a feature in Beta, which works as described and may still change.
• Partly it ships, and the answer states what is not supported yet.

Where it fits

• We already have an MQTT broker. Why would we need another layer on top?What a broker does not do, and what MaestroHub adds.Available
• Do we have to replace our OPC servers, SCADA and broker?Keep what you run; the one boundary on the namespace broker.Available
• Can one platform take data from PLCs, MES, maintenance and lab systems across all our shops?Connectors, one plant structure, a workspace per shop or domain.Available
• Which PLCs and protocols do you read? We have old lines and mixed brands.The protocol list, how to find tags, and what is not supported.Partly
• Every shop names and measures things differently. How do we make them comparable?Meanings and units, Start from your data, one topic translated on arrival.Beta
• Can we define assets, equipment types and processes, or only signals?What the plant structure and meanings cover today.Partly
• Do you follow ISA-95? Is it an OPC UA server? Sparkplug B?ISA-95 shape, OPC UA client only, Sparkplug B edge node.Partly
• If a value is converted to another unit, do we get a second topic?One topic, translated on arrival.Beta
• Can one alarm cover every similar machine?One limit on the meaning, later machines included.Beta

Security and IT approval

• Can it run entirely on our own servers, with nothing sent to you, even with no internet?Where it runs, the offline licence, and what telemetry holds.Available
• Will it pass our IT security review?Single sign-on, roles, encryption, TLS, and the MFA boundary.Partly
• Which certifications do you hold?ISO/IEC 27001 and how to check it; what we do not hold.Available
• Can it write back to our machines safely?Permissions, no late replay, approvals, and the boundary.Partly
• Can we limit each person to their own plant or line?Workspaces, access on a branch, Advanced Access Control.Available

Running it

• What architecture do you recommend for a plant-wide data foundation?Collect close, standardise once, deliver per use case; five layouts.Available
• We have many plants. Can we manage them centrally and roll one setup out to the next site?Workspaces per site, export and import bundles, what is not there yet.Partly
• What happens if the server dies?Replicas on Kubernetes, and what one machine does not cover.Partly
• Will it handle our volume?The dated benchmark, and sizing with you.Available
• What hardware and operating system does it need?System requirements and install options, Windows included.Available
• Do we need a box in every plant?Five layouts, edge or central, what does not replicate.Partly
• How do upgrades work, and can we go back?Upgrade check, apply and restore.Available
• Who runs it day to day, and will you train our team?Roles, operations, training; support terms per contract.Partly

Can we trust the numbers?

• How do you measure data quality?Three levels, who decides, frozen, silent and late signals.Available
• Can we trace a value back to the machine it came from?Lineage to the machine address, and the writer of every value.Available
• How do we know a data stream has stopped?Connection health, outage notifications, silent sensors.Available
• Something stopped working. Can we see who changed what, and undo it?The audit trail, its integrity check and its boundary; versions and revert.Available

Your stores and your systems

• Can we publish the standardised data to our time-series, vector and graph databases?Which stores have a connector, and the pattern for each.Available
• What happens to the data when one of those stores is down?A buffer on disk per destination, and delivery order.Available
• We already have a historian. Can we still report on last month or last year?The operational window, and long-term history in your store.Available
• We have no MES. How do you join SAP orders with machine data?The SAP routes, the join, and what is not supported.Partly
• Can we keep our own dashboards and BI?MQTT, REST, and writing to your BI database.Available
• Can you deliver to our cloud data platform in batches?Buffer by count or size, Parquet, the cloud targets.Partly

Process flow and graph

• Can we model our process flow and the relationships between machines?Publish the structure and the flow to your graph database today.Partly

Trying and buying

• Can we try it ourselves before we buy?The evaluation, the simulator, and the pilot shape.Available
• How is it licensed?Per site; editions and add-ons.Available
• You are a young company. What if you disappear, and who owns our data?Data ownership, open exports, signed releases.Available

AI agents

• Can our AI assistants use this data safely?MCP, one token per assistant, the same permissions as people.Available
• Can they use our own private model, or our Copilot?Any MCP client; the Copilot Studio guide.Partly

Asked something that is not here?

Ask us. Questions that come up more than once are added to this page.