A Combinatorial Algorithm for Fuzzy Parameter Estimation with Application to Uncertain Measurements

In this paper, we present a new method for the regression model prediction in an uncertain environment. in the practical engineering problems, to develop the regression or the artificial neural network model for making predictions, an average set of the repeated observed values is introduced to the model as an input variable. therefore, the estimated response of the process is also the average of a set of output values, where the variation around the mean is not determinate. however, to provide the unbiased and precise estimations, the predictions are required to be correct, on average, and the spread of date should be specified. in order to address this issue, we propose a method based on the fuzzy inference system, and the genetic and linear programming algorithms. we consider the crisp inputs and the symmetrical triangular fuzzy output. the proposed algorithm is applied to fit the fuzzy regression model. in addition, we apply a simulation example and a practical example in the field of machining process in order to assess the performance of the proposed method in dealing with the practical problems in which the output variables have the natures of uncertainty and impression. finally, we compare the performance of the proposed method with other available methods. based on the provided examples, the proposed method is verified for prediction. the results obtained show that the proposed method reduces the error values to a minimum level, and it is more accurate than the linear programming method and the fuzzy weights with linear programming method.