This article integrates fuzzy set theory in the data envelopment analysis (DEA) framework to compute technical efficiency scores when input and output data are imprecise. The underlying assumption in conventional DEA is that input and output data are measured with precision. However, production agriculture takes place in an uncertain environment, and, in some situations, input and output data may be imprecise. We present an approach of measuring efficiency when data are known to lie within specified intervals and empirically illustrate this approach using a group of 29 dairy producers in Pennsylvania. Compared to the conventional DEA scores that are point estimates, the computed fuzzy efficiency scores are interval bound allowing the decision maker to trace the performance of a decision‐making unit at different possibility levels.