Due to strong effect of Climate Change, the roads in Taiwan are not only prone to frailty, but also getting harsher to maintain in the recent year. Besides, the adversities such as the heavy traffic and the frequent road diggings challenged the maintaining engineers of the pavement management as well. While dealing with complicated situation involving the various decisions of the different authorities concerned in the engineering operations, the inexpert engineers depending on the technique and the experience merely passed verbally in the departments seem to be helpless. In the view of the situations, it is necessary to build an effective consulting system for engineers to breakthrough the predicament of pavement management. In this research, we have collected the methods of maintaining domestic urban roads and organized the models suiting for maintaining them. Then, build a semantic expert system about pavement maintenance knowledge providing appropriate maintaining policies by SWI-Prolog based on the information we have collected and organized. Last, we demonstrated the fruits how the expert system uplifted the pavement maintaining level.