In this paper, a data driven framework forperformance and maintenance evaluation (PAME) of windturbines (WT) is proposed. To develop the framework, SCADAdata of WTs are adopted and several parameters are carefullyselected to create a normal behavior model. This model which isbased on Neural Networks estimates operation of WT andaberrations are collected as deviations. Afterwards, in order tocapture patterns of deviations, self-organizing map is applied tocluster the deviations. From investigations on deviations andclustering results, a time-discrete finite state space Markov chainis built for mid-term operation and maintenance evaluation.With the purpose of performance and maintenance assessment,two anomaly indexes are defined and mathematically formulated.Moreover, Production Loss Profit is defined for PreventiveMaintenance efficiency assessment. By comparing the indexescalculated for 9 WTs, current performance and maintenancestrategies can be evaluated, and results demonstrate capabilityand effectiveness of the proposed framework.