Intersection signal timing is one of the key techniques in intelligent transportation system (ITS). Both the average delay and stop frequency are important indices for evaluating the level of service (LOS) for signalized intersections. Traditional signal timing models either optimize only one of them or deal with them as a single objective using weighted average methods. In this paper, a Multi-Objective Particle Swarm Optimization (MOPSO) method is proposed to optimize the both evaluation indices synchronously. A well-distributed set of Pareto optimal solutions is obtained, and the most satisfied solution is selected by the multi-objective decision-maker module. The experimental results indicate this optimal method is steady and effective.