The integration of stochastic simulation techniques into fuzzy comprehensive evaluation systems enables the derivation of probabilistic relative ranking outcomes, which is crucial for obtaining robust assessment conclusions. However, distinct stochastic simulation approaches are required for different types of fuzzy numbers, necessitating specialized research. Building on this foundation, this study focuses on the stochastic simulation aggregation problem for generalized trapezoidal fuzzy numbers (GTrFN). Firstly, the stochastic simulation method of generalized trapezoidal fuzzy number is studied. The random extraction method is determined according to the distribution function which is deduced by the membership function of the GTrFN. Secondly, by integrating the preference information of evaluators into the index clustering process, the index weight coefficient is determined, and the stochastic simulation density middle (SS-DM) operator of generalized trapezoidal fuzzy number is constructed to solve the evaluation results in the random aggregation process. Then, based on the determination method of the superiority degree of probability matrix, the rule of strong and weak superiority is introduced in the process of obtaining the ranking to reflect the difference among the evaluated objects in the ranking. Finally, an example is given to illustrate the effectiveness and characteristics of the proposed method.