This paper presents an adaptive neural network and fuzzy transform approach for coal consumed by the industrial sectors in different states which includes the 20 states and districts of Columbia and United States. The behavior of coal consumption is assessed based on the data of the time period 1960–2011. Despite of strength of these approaches, they have their own limitation too. The proposed algorithms are able to find a stochastic frontier based on a set of input–output observational data and explicit assumptions about the functional structure of the stochastic frontier are not needed. By applying these algorithms and calculating the efficiency scores, the ranking between decision-making units (DMUs) will be available. The superiority and advantages of adaptive neural network algorithm are shown by comparing its results against fuzzy transform algorithm.
