Determination of exact standard time with direct measurement procedures is particularly difficult in companies which do not have an adequate environment suitable for time measurement studies or which produce goods requiring complex production schedules. For these companies new and special measurement procedures need to be developed. In this study, a new time estimation method based on different robust algorithms of artificial neural networks (ANNs) is developed. For the proposed method, the products that have similar production processes were chosen from among the whole product range within the cleansing department of a molding company. While using ANNs, to train the network, some of the chosen products' standard time that had been previously measured is used to estimate the standard time of the remaining products. The different ANN algorithms are trained and four of them, which are converged the data, are stated and compared in different architectures. In this way, it is concluded that this estimation method could be applied accurately in many similar processes using the relevant algorithms.
A Novel Nonlinear Noise Power Estimation Method Based on Error Vector Correlation Function Using Artificial Neural Networks For Coherent Optical Fiber Transmission Systems