A Multi-Objective Optimization Model and a Decision-Making Method for Traffic Signal Control

The objective function plays a crucial role in traffic signal control systems. A new method for establishing the objective function is proposed, which uses the relative indices compared with a performance index of classical signal timing such as F-B. A Pareto-optimal front is obtained by solving a multi-objective optimization model with genetic algorithm, and the correlations of performance indices are analyzed. The most representative performance indices are selected as the final sub-objectives of the model. Furthermore, a fuzzy membership function is used to express the decision-maker’s preferences on the evaluation indices. Finally, the most satisfactory solution is obtained from the Pareto-optimal solution set. An isolated intersection is calculated as a case study, and the results illustrate that the proposed method expresses the decision-making preference of each objective more intuitively and makes the control target more clearly.