How to Automate AllMax Operator10

Operator10 is where the day's readings come to rest

A treatment plant produces one thing every day besides clean water, and that is numbers. Flows and levels off the instruments, chlorine residuals and turbidity from the bench, a round of gauge readings written on a clipboard, results on the samples pulled that morning, and a line in the log book about the pump that tripped at four.

Operator10 is where those come to rest. AllMax Software, part of the JDM Technology Group, sells it in a Water edition and a Wastewater edition, alongside Antero for maintenance work orders and a separate product for industrial pretreatment. The job it does is unglamorous and central: hold every reading a plant takes, calculate from it, and turn a month of them into the report the plant has to produce.

Operators run it, and so do superintendents and the lab. A plant that has been on it for years has a decade of daily numbers in one place, which is the only reason anybody can answer a question about last March.

Half the plant's numbers arrive as somebody else's PDF

Not every number a plant reports is a number the plant took.

The contract lab runs the analyses nobody does in house and posts them to its own results portal. A hauler brings septage and leaves a ticket. An industrial user files its self-monitoring results in whatever format it has always used, which is often a scan. The biosolids contractor's weights come off a spreadsheet emailed from a phone. Chemical deliveries of chlorine, polymer or ferric turn up against an order somebody placed three weeks ago and tracked nowhere.

So the last week of every month is a person with two screens open, copying figures across, checking them against the limits already entered, and finding out on the way that one result was never sent at all.

AllMax Operator10

Lab results matched to the sample Each analysis collected from the contract lab's portal as it posts and set against the sampling event already open for it, including the ones that never came back.

The month drafted rather than typed Readings, results and calculated values gathered into one package with the gaps listed, ready for the operator who reviews and signs it.

Deliveries and power read at source Chemical order status, hauler tickets and utility usage collected from the accounts that hold them, so a low day tank has a date on it.

The SCADA interface reaches everything inside the fence

Most of what the software is bought for is the part that already works, and a plant not using all of it should start there.

The OPC and DDE interface takes readings straight off SCADA and the Historian interface brings in what is already stored, so instrument data lands without anybody keying it. DataViews build custom views of raw or calculated values, and Sampling Events carry the detail attached to a set of lab results. Charts turn a column of readings into a trend and the Operator Log Book keeps the shift notes beside them. Permit limits can be entered so exceptions are calculated rather than spotted. Reporting ships with state forms and their built-in calculations, the Form Editor lets a plant reshape one, and data can be exported as XML for electronic submittal.

Every one of those starts from a value already inside the plant's own network. SCADA sits behind the fence. The contract lab's results portal, the hauler's account, the industrial user's own filing and the agency's submission site do not, and Operator10 was never given a login to any of them.

Lab results that find their own sampling event

Turn it around: the lab, the hauler and the utility all publish, and nobody is reading them on a schedule.

The contract lab's portal read each morning and each analysis matched to the sampling event already waiting for it, so a result stops being a file somebody has to find. Samples past their expected turnaround raised with the lab before the end of the month rather than during it. Hauler tickets and biosolids weights collected from the contractor's own account and totalled against what the plant recorded at the gate.

Then the things that decide next month. Chemical deliveries followed at the supplier, so a low day tank is a known date rather than a surprise. Power usage read off the utility account and set beside the million gallons treated, which is the number that moves a budget. Rainfall from a public gauge set beside the day's flow, so a wet weather spike explains itself. Parts for the pump that tripped tracked at the vendor until they have a date.

And the compiling itself. The month's readings, results and calculated values gathered into one draft with the gaps listed: which sample has no result, which day has no reading, which figure came back outside the limit already entered. What comes out is a package for review, not a submission.

An operator's licence is on the report, so an operator sends it

Assembling a month needs eight passwords and an hour, not a licence.

WebRun is an AI agent that works a real Chrome browser, signed in as the plant is. It opens the contract lab's portal, a hauler's account, the utility's billing page or a supplier's order status, reads what is there, and lines it up against what Operator10 already holds.

It runs on your schedule inside your own private environment, and a run can be limited to an explicit list of domains. Anything going to an agency, and anything that changes the plant's own record, is prepared and left for the certified operator whose name goes on it. Each workflow below names the sites it opens.