Engineering drawing man-hour forecasting based on BP-GA in design of chemical equipment

Abstract

The accurate man-hour forecasting is important in the control of project process and the optimization of human resources scheduling so as to cut the costs in the chemical plant design companies. This paper presents a framework, combining the Back Propagation (BP) Artificial Neural Network (ANN) and Genetic Algorithm (GA), for the forecast of Engineering Drawing Design Man-hour (EDDM), which is one of the most important work items in the design of chemical equipment. Based on the work flow analysis of chemical equipment design, the input variables are selected according to the result of contribution and correlation analysis. The data preprocessing and the forecasting model are also presented in details. Finally, the simulation results are discussed, which show that the model based on GA-BP ANN is better than the model based on the pure BP ANN, and the forecasting model based on GA-BP ANN is a helpful tool for EDDM forecasting.