The work aims to achieve the efficient distribution of fresh products, so as to improve customer satisfaction as much as possible. The characteristics of multimodal transport and cold chain logistics were analyzed to select the optimal logistics path. In the establishment of the path selection model, customer satisfaction was introduced as a parameter based on customer satisfaction with the quality and arrival time of fresh products. At the same time, in response to the national "carbon neutrality" policy call, consideration was given to the model construction. A low-carbon policy was adopted, with the objective function of minimizing total cost and maximizing customer satisfaction. With the cold chain transport of 15 node cities and three transport modes of highway, railway and water as an example, the relatively new Emperor Penguin Optimizer was compared with the traditional classic heuristic algorithm (Genetic Algorithm, Particle Swarm Algorithm) and the sensitivity analysis was conducted on the parameter of transshipment time. The customer satisfaction corresponding to the optimal intermodal path was 0.96 and the increase in transshipment time would lead to an increase in total costs, a decrease in customer satisfaction, and a change in the optimal path. The Emperor Penguin Optimizer has good convergence speed and convergence accuracy in solving multimodal transport problems, which provides a basis for the decision-making of multimodal transport carriers. The fluctuation of transshipment time can lead to changes in the fitness function value, so it is crucial to choose the appropriate transport method and route while considering customer satisfaction, and reduce unnecessary transshipment.