A rural revitalization strategy has been proposed, and more attention has been paid to village planning, which has become an important measure to adjust to local conditions, implement policies accurately, and build a beautiful countryside. The core of rural revitalization involves planning first, promoting industry, serving the countryside, and planning the village as the unit. An important aspect of village planning is road planning and design, which is essential to further connect agricultural production, country living, and ecological space of the village; promote production; serve life; and strengthen traffic between the village and the external region. Owing to the differences in geographical and natural conditions and the level of economic development between urban and rural areas in China, village road planning cannot copy the previous methods of road route selection and urban road planning. Instead, it is necessary to explore a universal, efficient, and intelligent village road planning method that considers the actual situation of rural development by considering agricultural production, villagers' lives, industrial development, historical and cultural protection, and other factors. In the context of rural revitalization, this study takes the village as the planning unit, considers the natural environment, current layout, and development needs of the village, uses the qualitative and quantitative fuzzy analytic hierarchy process to analyze and quantify the natural, social, demographic and other factors of the village road planning and their weights, and uses the weighted superposition analysis method to form a comprehensive cost grid of the village with multiple elements. A process-based village road planning GIS tool is constructed, and the minimum cost path algorithm was used to automatically generate the village road network with the minimum comprehensive cost. Taking Gangtou Village in Huadu District of Guangzhou City as an example, the feasibility of the above methods, technical processes, and GIS tools was verified, and a village-wide comprehensive cost grid was efficiently generated. According to the five elements of residential areas, cultivated land, historical monuments, public facilities, and entrances and exits of external roads, five categories, seven starting points, and 16 ending points of village planning roads were automatically extracted, and a four-level village road from multiple starting points to multiple ending points was automatically generated, including four main roads with a width of 5-8 meters, two secondary main roads with a width of 3-5 meters, nine productive branch roads with a width of 2-3 meters, 13 living branch roads, and one village road. The results show that the village road planning method proposed here can efficiently generate multi-type and multi-level village roads, and the planned road network not only has a minimum construction cost but also good connectivity and practicability. The village road-planning method proposed in this study integrates natural, social, demographic, and industrial development factors and their characteristics, providing systematic ideas and automated methods for village road planning. This is helpful for promoting the intelligence of village road planning.