Abstract: In this paper, the multi-objective optimization model of oil products distribution plan was established after the service features of oil products distribution and the influence factors of vehicle arrangement were analyzed. In the model, the traffic control policy of lorries in the downtown is taken as the constraint and the minimum transportation cost and risk are taken as the objective functions. According to the actual distribution situation of one oil company in Beijing, the adaptive simulated annealing genetic algorithm was designed to solve it. Then, the adjustability of the traffic control policy of lorries in the downtown which acts as the constraint was investigated. Moreover, the optimization suggestions based on the swapped distribution of different enterprises were proposed to address the irrational allocation of oil products resources. Finally, case verification was conducted from three schemes of existing conditions, constraint adjustment and swapped distribution. It is shown that under existing conditions, the minimum expense cost and risk cost to complete the same distribution task are RMB19 303 and 4 976 respectively. After constraints are adjusted, expense cost and risk cost decrease by 0.07% and 9.67% respectively. And when swapped distribution is carried out, expense cost and risk cost decrease by 11.87% and 38.34% respectively, and thus the distribution cost of oil products is cut down. The research results can be used as the reference for improving transportation safety of dangerous goods.