Abstract: In order to guide the foundry enterprises to scientifically select orders for production in the market, sand casting enterprises are the research object. Based on the comprehensive consideration of the evaluation indicators of the key departments of the enterprise, the order acceptance evaluation index system of sand casting enterprises is designed, and an order acceptance evaluation model based on hybrid GSA-GA neural network is constructed. In the algorithm, the Tent map is used to initialize the population to guarantee the random features, and the crossover and mutation operators of the genetic algorithm are introduced to maintain the global optimal particle acquisition and improve the algorithm exploration ability. To verify the validity of the model, an example of a sand casting company's order acceptance evaluation problem is used for experimental analysis, and the parameter experiment is carried out and compared with other algorithms. The results show that the average relative error of the model is 1.945%, which can help sand casting enterprises to make scientific order acceptance evaluation decisions and improve the production efficiency of sand casting enterprises.

Key words: GSA-GA algorithm, sand casting, neural network optimization, order acceptance evaluation