Automatic Health Assessment runs the same standard assessment on a machine on a schedule, records every run, and reports when something has changed. The result is a condition and a breakdown of the fault modes behind it, with the evidence attached, and a manager’s inbox that holds the machines whose situation moved rather than all of them.
A Condition, and What Is Behind It
The condition is deliberately short: acceptable, monitor, critical. Acceptable means carry on. Monitor means there is something worth watching but no action decided. Critical means a decision has been made and the evidence supports it.
Underneath that sits the part that makes it useful: the fault modes as a distribution that adds to a hundred. “70% unbalance, 25% bearing outer race, 5% other” tells a planner what the job is. A healthy machine reads as healthy at a hundred percent, which is a result in its own right.
Silence Is Information, and Spam Is Not
The decision that shapes everything else is what happens when nothing has changed. A routine assessment that finds nothing new leaves a short record that it ran, and sends nothing at all.
A report is produced, the machine’s health status is updated and the people responsible are emailed when the condition changes, when a new fault mode appears even though the condition has not moved, and when a machine comes back to health. Good news gets reported too. Percentages shifting within the same set of fault modes do not raise anything.
Assess two hundred machines in a week and what reaches a manager is the handful whose situation actually moved.
Every Run on the Record
Each assessment is written to a structured history: when it ran, what condition it found, the fault distribution, a summary of the evidence, and how it differs from the assessment before it. That history is what turns a set of judgements into something you can audit and plan against, and it answers the questions that are usually guesswork: when was this machine last assessed, which ones have been left for three weeks, which ones got worse this month.
It is also what makes a change detectable at all. A change is a comparison, and a comparison needs something to compare against.
The Decision Stays Yours
An assessment can propose an action, and a proposal arrives with the evidence behind it. Raising the work order remains a human decision. The value is not that the platform acts on its own; it is that the person who decides gets the case already made, on every machine, every week.
Frequently asked questions
What does an Automatic Health Assessment produce?
Two things. A condition for the machine, on a deliberately short scale of acceptable, monitor and critical; and a breakdown of the fault modes behind it as percentages that add up to a hundred, so a machine reads as "70% unbalance, 25% bearing outer race, 5% other" rather than as one word. Each finding carries the evidence it rests on.
Will it fill my inbox?
No, and that is the design. A routine assessment that finds nothing new leaves a short record and sends nothing. A report and an email go out when the condition changes, when a new fault mode appears even at the same condition, and when a machine returns to health. Percentages moving inside the same set of modes do not trigger anything.
Does it open work orders by itself?
No. The assessment can propose an action, but raising the work order stays a human decision. What it does do is put the proposal in front of the right person with the evidence attached, so the decision takes a minute instead of an afternoon of chart reading.
Can I see when a machine was last assessed?
Yes. Every run is written to a structured history, including the ones that changed nothing, so questions like "when was this machine last assessed", "which machines have not been looked at for three weeks" and "which ones got worse this month" have answers rather than estimates.
Why keep a record of an assessment that found nothing?
Because silence means two very different things. A machine nobody has assessed and a machine assessed yesterday with nothing found look identical in an inbox and are not remotely the same situation. Recording the quiet runs is what separates them, and it is also what makes a change detectable, since a change is only meaningful against a previous assessment.