Summary: In this paper we develop an approach that synthesizes the best features of the two main methods in the estimation of production efficiency. Specifically, our approach first allows for statistical noise, similar to stochastic frontier analysis, and second, it allows modeling multiple-inputs-multiple-outputs technologies without imposing parametric assumptions on production relationship, similar to what is done in non-parametric methods. The methodology is based on the theory of local maximum likelihood estimation and extends recent works of S. C. Kumbhakar et al. [J. Econom. 137, No. 1, 1–27 (2007; Zbl 1360.62131)]. We will use local-spherical coordinate system to transform multi-input multi-output data to more flexible system which we can use in our approach. We also illustrate the performance of our approach with simulated example.