To investigate the impact of various factors on total cost and carbon emissions in different transportation modes, we propose a green vehicle routing problem model incorporating simultaneous split delivery and pickup. The model aims to identify environmentally friendly routes while minimizing overall costs by integrating approximate calculations of fuel consumption and carbon emissions. To tackle this issue, we have developed a two-stage approach. In the first stage, we utilize two clustering methods to group customer domains in scenarios with heterogeneous and homogeneous fleets. Subsequently, within each cluster, we determine suitable routes based on the characteristics of goods delivery and pickup at each node to accommodate load fluctuations. In the second stage, we employ the CMEM fuel consumption model to calculate the total cost. Through numerical experiments with the Solomon benchmark dataset, we validate the effectiveness and feasibility of our proposed model and algorithms. The results indicate that factors such as vehicle speed, load rate, driving distance, and number of routes have varying impacts on fuel consumption and carbon emissions. Although carbon emissions account for a small proportion of total cost implications, it is recommended to introduce new energy vehicles into the transportation market. Furthermore, the use of heterogeneous vehicles can enhance cluster flexibility by reducing the number of required transport vehicles and routes, leading to significant reductions in fuel consumption, carbon emissions, and overall transport costs.