Aiming at the problems of high cost, large carbon emissions and low customer satisfaction in the cold chain distribution of perishable goods, this paper measures customer satisfaction from the timeliness and quality of perishable goods distribution and takes this as a constraint. Considering the fixed cost, transportation cost, cargo damage cost, refrigeration cost, penalty cost and carbon emission cost in the distribution process, this paper constructs a vehicle routing optimization model of perishable goods cold chain distribution with the minimum total cost as the goal, designs an improved genetic algorithm to solve the optimization model, and analyzes the complexity of the algorithm. The results of numerical experiments show that the designed algorithm can always obtain a distribution plan with lower total cost, higher product freshness and less carbon emissions. They also show that the improved genetic algorithm has certain advantages in cost saving and customer satisfaction improvement compared with the traditional genetic algorithm, and also verify the rationality of the model and the effectiveness of the algorithm to some extent.