Aiming at the high cost of carbon emissions in the logistics of e-commerce platforms and the uneven benefit of the delivery staff during the distribution process, in order to meet the needs of reducing transportation and labor costs for the e-commerce platform, improve vehicle delivery efficiency, reduce carbon emissions, and achieve low-carbon green travel. A delivery benefit balance-based green vehicle routing problem with time windows and its intelligent solving algorithm are studied in this paper. Firstly, a multi-objective green vehicle routing model is established by designing the functions of fuel consumption based on driving speed, penalty cost on fuzzy customer satisfaction, and the variance of the delivery benefits, with the goal of minimizing fuel consumption, penalty cost, and distribution revenue variance. Then, a multi-objective evolutionary algorithm is developed to solve the above model by integrating variable neighborhood search operators into NSGA-II. Finally, eighteen benchmark data-sets are selected from Solomon test data-sets for experiments, the comparison with two models and three algorithms on the hypervolume value and the knee point value validates the feasibility of the proposed model and the effectiveness of the solving algorithm, which can provide new solutions to reduce carbon emissions and realise low-carbon green travel.