At present, the financial crisis is the biggest crisis faced by listed state-owned Enterprise in China, and thus the establishment of early warning model of financial crisis is becoming more and more important. Genetic algorithm is more and more widely used in enterprise financial crisis. At present, many scholars have designed a part of early warning model based on genetic algorithm. In this paper, through the optimization and improvement of genetic algorithm, an enterprise financial crisis early warning model optimized by genetic algorithm was reestablished, so as to improve the enterprise’s ability to resist risks and predict the financial crisis ahead of time, and help enterprises cope with the difficulties. The test shows that the model is safe and reliable, and can accurately predict the financial crisis of the enterprise.