Many real-world optimization problems have to face a lot of difficulties: they are often characterized by large and complex search spaces, multiple conflicting objective functions, and a host of uncertainties that have to be taken into account. Metaheuristics are natural candidates to solve those problems and make them preferable to classical optimization methods. We here propose a number of new evolutionary algorithms to find a set of non-dominated solutions from multiobjective optimization problems in uncertain environments. Experiments are conducted on multiobjective scheduling with stochastic processing times.