In this paper, we propose a support method for multi-objective layout design of cellular manufacturing systems. The proposed method is composed of a global search technique for layout designs and a data mining technique applied to the obtained layout solutions. First, intercell layout design candidates (possible locations of each manufacturing cell or department within a floor area) are encoded as sequence variables, using a sequence-pair scheme, and a multi-objective genetic algorithm using NSGA-II is then used to search the set of Pareto optimal solutions for the intercell layout design problem. Then, to analyze the solution set and simplify the subsequent decision process, we develop a data mining technique, EM clustering, which clusters the obtained layout solutions considering correlations of cells or departments among the layout solutions. To demonstrate the effectiveness of the proposed method, we present a numerical example of a two-objective intercell layout design problem that aims to minimize the material handling cost and maximize the closeness rating.