Utilities often call in mutual aid crews to restore power after severe storms, but they face a timing dilemma. Requesting crews early increases the chance of securing enough workers, yet early decisions depend on weather forecasts that are less reliable the further out they reach. A new preprint on arXiv examines this trade-off directly, asking how forecast lead time changes the practical value of outage predictions.
The study uses AI weather models to simulate outage forecasts at different lead times and evaluates how those forecasts translate into crew deployment decisions. The authors quantify the point at which the benefit of securing crews earlier is offset by the cost of acting on less accurate information.
While the abstract does not report specific numbers, the framework offers a way for utilities to weigh early resource acquisition against forecast uncertainty. The work is presented as a cross-listed preprint, meaning it has not yet undergone peer review, but it points toward more data-driven scheduling for storm response.