The assumed method could be based on the time series of nominal production and the correlation of real production to the number of sunny days in individual months and years ... any idea?
Dear Mr. Adam Pawliczek, I suggest Dynamic DEA, which enables you to compare the efficiency of each photovoltaic panel over time, based on input [resources] and output [outcomes, products] variables.
Reference:
Tone, K., & Tsutsui, M. (2014). Dynamic DEA with network structure: A slacks-based measure approach. Omega, 42(1), 124-131.