Sensitivity analysis of the order economic value model in the case of two-factor interactive effects using artificial neural network

In this article, how to use the artificial neural network metamodel to analyze the sensitivity of the order economic value model in the case of two-factor interactive effects is presented and it is shown that the use of this metamodel for the sensitivity analysis of the economic value of the order compared to the current method (one factor in each load) is more suitable. For this purpose, a back-propagation forward neural network with a hidden layer, sigmoid driving functions in the hidden layer, factors affecting EOQ as input and economic order value as output of the model have been used. The sensitivity criteria are defined based on connection weights and the methodology procedure is shown in a numerical example. With the emergence of intelligent systems, data processing and models related to them such as artificial neural networks, genetic algorithm, fuzzy logic and the like, which are designed and modeled with inspiration from a corner of nature, an important progress has been made in data analysis. One of the main features of the network Artificial neural networks, inspired by the structure and function of natural neural networks, are parallel processing of input information by neural processing units.