Fuzzy data envelopment analysis (DEA) has been widely applied in uncertain input-output system evaluation involving fuzzy data. Since the existing fuzzy self-evaluation DEA models usually lead to overestimation of efficiency, this paper proposes a novel fuzzy cross DEA evaluation model that integrates fuzzy possibilistic mean and prospect theory. First, triangular fuzzy input-output data are converted into interval numbers based on the lower and upper possibilistic mean, and an improved interval cross-efficiency model is constructed. Second, prospect theory is introduced to dynamically model decision-makers’ psychological preferences and gain-loss of cross efficiency with respect to self-evaluation efficiency. By maximizing the prospect cross-efficiency of each decision making unit (DMU), we get the optimal aggregation weights and evaluate the prospect-weighted cross-efficiency interval of each DMU. Then, a preference degree method of cross efficiency interval is utilized to rank all the DMUs. One numerical example is given to demonstrate the effectiveness and advantages of the proposed fuzzy cross-efficiency DEA model.