Objective: To optimize extraction process of total flavonoids from Nervilia fordii by central composite design and response surface methodology. Method: Central composite design was used with the concentration of ethanol

extraction time and the amount of ethanol as independent variables

the content of total flavonoids as dependent variables

each level was done with multiple linear regression and binomial fitting.Technology parameters was optimized by response surface methodology

and predictive analysis was carried out by comparing observed and predicted value. Result: Established mathematical model had high multiple correlation coefficients

relationship between factors and indexes was suitable for fitting with binomial equation(R2=0.987 0)

verification test result and model prediction value had small deviation.Optimum extraction process was:extracted 2 times with 30 times the amount of 65% ethanol

75 min per time. Conclusion: This optimized process was simple

and had well prediction

it was suitable for development and utilization of total flavonoids from N. fordii.