Optimizing Back Propagation Neural Network with Genetic Algorithm for Man-hour Prediction in Chemical Equipment Design

Abstract

This papers sets to address the low accuracy of man-hour prediction, and proposes modeling based on optimizing back propagation neural network with genetic algorithm (AG_BP) for quantitative predictions. We conduct research on the management process of chemical equipment design, based on historic data in the user database, analyze the relevance of parameters and obtain the parameters, construct and improve models using the AG_BP algorithm. The results show that this approach is a good solution for predicting man- hours required for chemical equipment design. The models designed in this paper can help improve the prediction accuracy and can be promoted for broader use.