The　arithmetic　average　method　is　usually　adopted　to　aggregate　cross-efficiency　in　traditional　cross-efficiency　methods.　However,　this　method　not　only　underestimates　the　importance　of　self-evaluation,　but　also　ignores　the　subjective　preference　of　decision-makers.　This　paper　thus　introduces　prospect　theory　to　describe　the　subjective　preference　of　decision-makers　in　the　aggregation　process　when　they　face　gains　and　losses,　then　a　new　method　is　constructed　to　aggregate　cross-efficiency.　Based　on　the　differences　between　the　psychological　expectations　and　aggregation　results,　the　expectations　are　constantly　adjusted　until　a　consensus　on　aggregation　results　is　reached.　An　aggregation　result　that　is　more　acceptable　to　all　decision-making　units　can　then　be　obtained.　Finally,　the　proposed　method　is　applied　to　aggregate　the　cross-efficiency　of　27　industrial　robots　to　illustrate　its　effectiveness　and　convergence.　©　2019,　Springer　Science+Business　Media,　LLC,　part　of　Springer　Nature.