In this paper, on the background of Internet of Things (IoT) technology, a three-level logistics network consisting of multiple suppliers, manufacturers, and distribution centers is established, and multiple decision-making problems of material collection, logistics generation, transportation modes, and transportation routes in the logistics network are solved by setting up a mixed-integer nonlinear planning model that minimizes the operation cost and carbon emission cost of the logistics network. After that, the optimization is carried out by max-min ant system and ant colony algorithm, so that the improved ant colony algorithm sets the solution equations of the specific model targeted to the specific problems. The results show that the algorithm in this paper can effectively optimize the logistics distribution problem, and its transportation distance is significantly reduced (6.38%) compared with the traditional ant colony algorithm, and the algorithm in this paper can solve the logistics path optimization problem faster, which controls the cost of transportation to a certain extent. Under the multiple conflicting objectives of simultaneously considering transportation cost, carbon emission and cargo loss rate, the government can increase the market share of railroad cold chain transportation by giving appropriate tariff subsidy policy or by increasing the travel speed of railroad cold chain liner. In enhancing the competitiveness of railroad cold chain logistics transportation, the tariff subsidy and the cold chain train can be substituted to a certain extent.
