This thesis proposes a new method to fin the Pareto optimum solutions of the multi-objective topology optimization problem. Conflicting objective functions are converted to a single objective function by applying weights, and these weights are adaptively updated to find evenly spaced solutions on the Pareto front.