Nowadays, the road traffic in Taiwan is getting more and more busy, and the mastery and arrangement of time has become a topic that the government and the people attach great importance to. In order to effectively master and arrange travel time, we propose a travel time prediction system that includes front-end operation interface and back-end data analysis. The front-end operation interface has two modules. The first one is the time estimation module. Its function is to serve as a bridge between the user and the database. The second is the LineBOT module. Its function is to store the data of the database. Calculated as travel time. The back-end data analysis system is divided into three modules: First, we design the data filtering module to organize and reduce the Open Data data, and then use the data clustering module to analyze the data to obtain smooth data, and finally use the time backtracking. The module converts the data into travel time and saves it to the database for the user to query. Experiments have shown that our proposed system can effectively predict travel time.