Abstract: A reasonable time-of-use price policy for natural gas is conducive to the resource allocation, and could guarantee the stable operation of natural gas network, so that the “gas famine” can be relieved, which is helpful for the healthy development of gas industry. Therefore, on the basis of users' response behavior, this paper constructs the multi-objective time-of-use pricing model, with minimizing the maximum peak load as well as difference between peak load and valley load as the objective, price restrictions obtained from gas enterprise revenue, user revenue and government as the constraints. Integration of genetic algorithm and Pareto ranking is adopted to solve the model. Finally, in a case experiment, the study divides 24 hours in a day into three different times, i.e. flat, valley and peak, and assumes the rule of“large amount, high price”. The computational results show the existence of an optimal price ratio for peak, flat and valley, respectively, from which both the gas company and customer can benefit at the same time, and that the difference of peak load and valley load can be reduced effectively, which verified the effectiveness and feasibility of the multi-objective model.