Scheduled maintenance is planned against a calendar and unplanned maintenance is planned against nothing. What determines a fleet's availability is the work in between: the interventions that could have been scheduled had the failure been known about early enough.
The repair is the same. The mobilization, the lost production, and the collateral damage are not.
Alerts are ordered by the probability that degradation continues, the production at risk, and the cost of the intervention, so the schedule reflects the condition of the fleet rather than the anniversary of its commissioning.
Several weeks of lead time is what allows work on different machines to be consolidated into one mobilization. That is the whole economics of offshore maintenance and a large part of it onshore.
A component identified six weeks out is a parts order and a crane booking. The same component identified on the day of failure is an outage of unknown length while both are arranged.
A crew sent to a healthy machine spends the same money as a real intervention. Thresholds are set against what a visit costs, which differs by an order of magnitude between an onshore site and a vessel charter.
The work a failing component requires does not change with the date it is discovered. What changes is the mobilization, the production lost while the machine waits for parts nobody ordered, and whether the failure was given time to damage anything else.
That is the entire argument for early detection, and it is why lead time rather than raw accuracy is the measure that matters most. An alert that is correct on the day of failure has told the team nothing it was not about to find out, and it has spent the team's confidence to do it.
Access windows, crews, cranes, vessels, and parts lead times are all constrained, and a planning team solves that problem well when its inputs are good. The input usually missing is a reliable statement of which machines will require attention in the next quarter and what each one is worth.
With that input the plan changes shape. Work is grouped by location and access rather than by the order it was reported. Low-value interventions are deferred deliberately rather than by omission. The campaign that requires a crane carries everything that needs one, instead of the crane returning twice in a season.
Work orders and service history are the ground truth the analysis depends on. A machine whose gearbox was replaced appears as an unexplained change in behavior unless the replacement is known, and a control-side correction applied after a finding is what confirms the finding was right.
That feedback loop is the part of the platform that improves with use, and it is why maintenance records are worth connecting even though a first assessment does not require them.
The objective is not fewer repairs. It is a larger share of the same repairs happening on a date the team chose.
See how OpenTurbine can help your team understand performance, anticipate operational issues, and make better decisions across your wind assets.