Abstract:Since the selfdriving has many advantages, such as liberalization, customization, flexibility and comfort, now many travel enthusiasts have been attracted by selfdriving. And the quality of the tourist routes may directly affect the satisfaction of selfdriving tourists. In this paper, the national 5A tourist attractions are taken as the destination, a selfdriving tourist in Xi’an City is selected as the research object. This article uses modified circle algorithm to design the tourist route of each province, sets the optimization of time and cost as objective function, takes the limit value of travel time per year and the limit value of single travel time as constraint condition, and builds the integerlinear programming model. The optimal selfdriving tour decision of the national 5A tourist attractions is solved with LINGO software, which has a certain applicability.