The invention provides a method for classifying rock mass quality through normal distribution fitting rock mass wave velocities. The method comprises the following steps of (1) ranking measured rock mass wave velocity data from small to large, counting the number of the wave velocity data, and normalizing wave velocity data probabilities; (2) utilizing a normal distribution probability density function under a least square method for fitting the wave velocity data with the probabilities being normalized, and obtaining a rock mass wave velocity data model function; (3) utilizing the characteristic that a normal distribution probability density function f (x, mu, sigma) curve is an inflection point at a part with x being equal to mu+/-sigma, adopting x being equal to mu i+/-sigma i as a rock mass classification standard value in a rock mass wave velocity data model function F (x), and utilizing an expected value and a standard deviation in the rock mass wave velocity data model function F (x) for carrying out rock mass classification. According to the method provided by the invention, the rock mass quality is classified by utilizing the normal distributionprobability density function in statistics according to the work arearock masswave velocity values, so that the workload of a field test at a construction site can be reduced, and the work efficiency is remarkably improved.