The answers you normally have to dig out of e-mail threads

The MES is a product that collects data on the floor at the moment events happen: orders, stoppages, scrap, reports. A Power BI integration is a service — it takes data that already exists somewhere (in the MES, ERP, SCADA or files), arranges it into one model and builds reports from it. The MES answers “what is happening on the floor right now”, Power BI answers “what did it look like over a month and a quarter”.

No. We start from what already works — controllers, SCADA, MES, ERP, even spreadsheets. The data layer is built alongside them, not instead of them, so no existing system changes the way it works because of this implementation.

No. We deliver Power BI integrations for plants that already run a different MES, and we implement HyperVision without BI reports. They work best together, because the reporting data module feeds the model straight from shop-floor events.

The data model, the reports and the technical documentation — together with the knowledge of how to maintain them. Team training is part of the scope precisely so that changing a colour on a chart does not require a ticket to us. Post-implementation support, if it is needed, is a separate, explicit item rather than a condition for the reports to work.

The amount on the Pilot package is an entry threshold — it covers one area and one data source. Standard and Scale are quoted individually, because their scope depends on the number of sources and on the state of the data. We give the specific figure after the workshop, before the work starts.

One business topic for one audience — line OEE for shift supervisors, for example — across at most three tabs of a single Power BI file built on a shared data model. A new topic or a new audience is already another report. Changing the layout, the colours or adding a filter is part of the package; the full boundary is set out on the pricing page.

A station is a point where production is recorded — an operator panel, a machine, an inspection bench. The unit rate falls as the number of stations covered by the system grows, according to the thresholds given in the pricing.

Yes, that is the most common scenario. We start with one line and extend the scope after the first quarter — which usually brings a lower price threshold into play.

Not within an existing configuration. Every module needs its own integration and database preparation, so the set is fixed once, at the start of the implementation. Adding a new module later is a separate, smaller integration project — quoted just as openly as the first implementation.

The amount above covers the system itself, in the subscription model. Connecting machines and company systems is a separate stage: it starts with an integration audit and is quoted separately — the scope depends on what stands on the floor and on what the ERP exposes.

It depends on the system. Power BI runs in the Microsoft cloud, but on the plant’s own tenant — not on our infrastructure — and the data layer can be kept entirely on company servers. The HyperVision MES is always deployed on-premise, inside the plant’s closed network — with no data leaving the site and no cloud variant.

That is the starting position in most plants and a normal part of the work. The kick-off workshop exists exactly to establish what can be calculated today, what is missing and what has to be measured. We come out of it with a list of metrics that can be calculated from what already exists.

Yes, that is the most common case. Machines are retrofitted with IoT sensors — vibration, temperature, cycle counter — and read independently of the age of the controller. The age of a machine decides how it is measured, not whether it can be covered by the system.

Access is set at row level (RLS), so the head of one department does not see the results of the next one, while the board sees the whole picture. In the MES the same split is handled by roles: an operator, a maintenance technician, a shift supervisor and a manager get different modules and different buttons. The scope of permissions is agreed at the workshop, before go-live.

The first working report on plant data takes 2–3 weeks — that is the scope of the Pilot package. Not a presentation and not a mock-up, but a report that can be opened in the morning before the stand-up. Only then is the decision made about going further.

The kick-off workshop takes one day. After that we need access to the sources and roughly 2–4 hours a week from somebody who knows the process and knows where things sit. Installation, the model and report maintenance are on our side, and the full split of tasks is written out on the implementation page so nobody discovers their role halfway through.

Then we stop at the pilot, and the plant gets that in writing along with a recommendation on what to change in how data is collected. That is why we start with a package from 8,000 PLN rather than with rolling out the whole plant — the risk of the first step has to be one you can carry.