With the continuous development and growth of the transportation industry, people have higher requirements for the quality of transportation. Among them, the expensive cost of transporting dangerous goods is unacceptable. In response to this problem, this paper proposes a multi-objective optimization algorithm to reduce transportation costs and improve the efficiency of goods transportation. This article uses multi-objective evolutionary algorithm as an optimization tool, combined with key indicators of dangerous goods transportation, including transportation cost, dangerous goods leakage probability, and transportation time, to construct an adaptive evaluation model. By simulating different transportation scenarios in the experimental environment, the multi-objective evolutionary algorithm and particle swarm optimization algorithm are compared and analyzed. The experimental results show that the total transportation cost of the method proposed in this article is between 30,000 and 60,000 yuan. The optimization method based on multi-objective evolutionary algorithm has achieved significant results in the railway public transportation service network for dangerous goods.