It is well known that the performance of any evolutionary multiobjective optimization (EMO) algorithm over one class of problems is offset by the performance over another class by the “no free lunch” theorem. This means that there is no EMO algorithm can be regards as a panacea. Therefore, we propose an evolutionary multiobjective optimization algorithm with algorithm adaptive selection. It divides the population into several small subpopulations according to their distribution in the objective space. Each subpopulation owns a EMO algorithm, and make the worst agent on specific measures of performance learn from its neighbor best one according to the feedback from the search process. We test the proposed algorithm on nine widely used test instances. Experimental results have shown that the proposed algorithm is very competitive.