Product quality prediction based on RBF optimized by firefly algorithm

With the development of information technology, a large number product quality data in entire manufacturing process is accumulated, but it not explored and used effectively. The traditional prediction models have many disadvantages, such as high complexity low accuracy. To overcome above problems, we propose an optimized equalization method to pre-process dataset design simple effective model: radial basis function model by firefly algorithm with Levy flight mechanism (RBFFALM). First, new introduced dataset, which reduces dimension data, removes redundant features, improves distribution. Then RBFFALFM predict quality. Comprehensive expe riments conducted on real-world datasets validate that combining pre-processing outperforms other previous me thods predicting