The basic assumption in data envelopment analysis patterns (DEA) (such as the CCR AND BCC models) is that the value of data related to the INPUTS AND OUTPUTS is a precise AND positive number, but most of the time in real conditions of business, determining precise numerical value is not possible in for some INPUTS or OUTPUTS. For this purpose, different models have been proposed in DEA for imprecise data over recent years AND also several researches have been conducted on DEA that are able to evaluate efficiency with negative data. The negative INTERVAL DEA pattern which has been introduced AND used in the present study, addresses uncertainty both in INPUTS AND OUTPUTS AND provides user with more stable AND reliable results for decision making.Now, in this paper a model is presented that is able to compute efficiency INTERVAL of units with INTERVAL input AND output that while some indicators can also be negative AND then we prove that the efficiency INTERVAL that this model gives us is more precise compared to efficiency INTERVAL of models previously proposed AND finally, ten decision making units (DMUs) with the negative imprecise (INTERVAL) data are investigated by the proposed model AND the results of the proposed model are compared with the results of the previous models.