Process industry
At a process plant the question is rarely what is happening right now. It is what happened back then: what kind of run produced the batch that deviated, how much energy went into each tonne produced and what changed last month. Answering it requires that the measurements are stored far enough back and can be retrieved by time range.
Short answer
DataPortia™ stores process measurements as a time series on your own server and produces reports over a chosen time range — the run of a single batch, for example. Specific consumption is calculated with calculation points, and history is retained for 24 months by default. Recipes and quality data never leave the plant network.
How is a batch report produced?
By selecting a time range. DataPortia stores measurements as a continuous time series, and the report is built over the range that is selected — in practice from the start of the batch run to its end. Aggregates are calculated over the same range: minimum, maximum, average and sum. The same report template works both for a report run by hand and for a scheduled one.
DataPortia does not read batch IDs from an MES and does not identify batches automatically. A report is bounded by a time range. If batch starts and ends are to be picked up automatically, they have to come from the automation system as a measurable signal — tell me the situation and I will go through how it is done in practice.
| What is reported | Where it comes from | The part of DataPortia |
|---|---|---|
| Batch running conditions | Temperatures, pressures and flows during the run | Time series + a report over a chosen time range |
| Specific consumption | Energy in relation to the quantity produced | Calculation points that adapt to the time range |
| Background of a quality deviation | The same measurement from the past | History trends, zoom and presets |
| Process disturbances | Alarms and events | Original ms timestamp, Pareto analysis |
| Monthly follow-up | The same measurements over a longer period | Scheduled PDF, CSV or Excel report |
How do I find out afterwards what lay behind a quality deviation?
By looking at the same measurement from the moment the deviation arose. That requires two things: that the data is stored far enough back, and that you can zoom into a single period within it. The retention period is adjustable and 24 months by default, and history trends can be zoomed and saved as presets for reviews you repeat.
In detail: retention, compression and what they mean for retrieval
Automatic compression starts after two days and saves over 90 % of the disk space. Compression does not limit the time range data can be retrieved from — only the retention period does. In practice a long history is therefore cheaper than the disk usage of the first days would suggest.
Set the retention period according to how far back a quality investigation may need to reach. If a customer complaint can arrive a year after delivery, 24 months is a different matter from three months. Details on the OPC UA data acquisition page.
How is specific consumption calculated?
With a calculation point. It is a formula-based derived measurement point built from measured values — from power into energy, for example, and from there into a ratio against the quantity produced. The point adapts to the time range of the report, so the same definition produces the batch, the day and the month figure without a separate calculation.
In detail: why the calculation is worth defining once
When the same figure is recalculated in a spreadsheet every time, the method starts to drift between reports — a different person bounds the period differently or uses a different coefficient. A calculation point is defined once, and every report uses the same definition.
This is also the point at which a reporting system starts to pay for itself: the manual work goes away and the figures are comparable from one period to the next.
Does process data stay on the plant network?
It does. DataPortia is installed on the plant's own Windows or Linux server, and data acquisition, the database, reporting and AI analysis all run there. No internet connection is needed. In the process industry this is often decisive: running conditions are a trade secret, and nobody wants them in an outside cloud service.
In detail: what is handed out, and on what terms
Data moves outwards only in the way separately agreed. REST API v1 is read-only and authenticates with an API key, which is allowed 120 requests per minute. Webhooks push events to the receiving system, HMAC-signed. Neither writes to the automation system.
If the recipient is an MES or an ERP, give each integration its own key — then load can be followed system by system, and revoking one key leaves the others untouched. See REST API and webhooks.
Frequently asked questions
Below are the questions that come up most often at process plants: how many measurement points the system takes, what has to be done to the automation system, whether old measurements can be brought along and how data moves into an MES. If your requirements list has something this page does not answer, send the list — I will answer it in writing.
How many measurement points can the system take?
The limiting factor is throughput, not the number of points. DataPortia takes in over 2 000 values per second, which is about 172 million rows per day. How many points that is enough for depends on the sampling rate — points read rarely fit in far greater numbers than points read frequently.
Does the automation system have to be modified?
No. DataPortia connects to the existing server as an OPC UA client and reads from it. The PLC program is not touched and no software is installed on the automation side. What is needed is a network connection to the server and credentials with read rights.
Can old measurements be brought along at commissioning?
Yes, if the OPC UA server has a historical archive in use. The history import reads the past period from there and skips the parts already imported, so the same period can be run again without duplicates. Without a historical archive, collection starts from the moment of installation.
Can the reports be run automatically every month?
Yes. Reports are scheduled on a daily, weekly or monthly rhythm and are produced as PDF, CSV or Excel. The report template defines the logo, cover page, page size and signature blocks, so the output looks like the plant's own document as it comes. See Automated reports.
How do I test this at my own plant?
Install the trial version on your own server and connect it to a real OPC UA server. The trial runs 30 days, with every feature except the HA add-on, and no commitment. Pick one real batch and produce a report from it — that tells you faster than any demonstration whether the points you need come through and whether the output will do as it stands.