Abstract: Generalized exponential distribution is one kind of commonly used distribution which finds wide applications in reliable field. In recent years, the researches focus mainly on the methods of statistical analysis for treating common life test data. Literature for studying parameter estimation involving grouped data is rarely seen. In view of this, an exploration is made into the parameter estimation problem of generalized exponential distribution in the constant-stress accelerated life tests which are based on grouped data. Then, the two-parameter concrete expressions were given by use of EM algorithm. At last, the Monte Carlo stimulation proves the efficacy and feasibility of the estimation method