Abstract: Purpose/Significance Multi-trip delivery allows the vehicle to travel multiple times between the distribution center and the distribution point, which has the advantages of high efficiency and low vehicle dispatch cost. However, given the changes of the delivery time and times of delivery due to multiple round trips, customers have generated different psychological feelings of the service. Meanwhile, it is more difficult to optimize multi-trip routes considering the customer satisfaction. Design/Methodology In response to the requirements of customers on service time and the service times, this paper measures the customer satisfaction by building satisfaction functions based on delivery time window and service times, and combines with the stochastic uncertainty of the pickup-delivery demand. The multi-objective optimization model for multi-trip delivery is established to maximize customer satisfaction and minimize transportation cost. Considering stochastic and simultaneous pickup-delivery characteristics of demand in the model, the stochastic chance-constrained programming is introduced. A real-time adjustment strategy of “point judgment” is put forward, to design a grey relational multi-objective tabu search algorithm. Conclusions/Findings As a result, an example shows that the optimization model and algorithm are feasible and effective.