Advanced Traveler Information Systems (ATIS) aim at collecting and disseminating helpful information to travelers. Among them there are systems providing public transit information as well as path planning services, called Advanced Public Transportation Systems (APTS). However, most modern APTSs maintained their own individual transportation service information in a non-sharable fashion, thus planning paths for large scope is difficult since there is no information exchanging methods for existing APTS. In addition, existing researches usually employ shortest path algorithms, and did not take advantages of other semantic information such as geographic locations, service information, etc., thus usually are not scalable well for large scope trip planning. This work proposes a Semantic-based APTS framework for transit planning, where Semantic Web-based service interfaces are specified for information exchanging, making it possible to integrate and organize heterogeneous data from various APTS and transportation enterprises. In addition, a semantic-based trip-planning scheme is designed to be used in NimbleTransit, where service information is aggregated in advance using semantic knowledge. Semantic service composition is then used to construct the path. A system prototype is implemented, and the results showed that the proposed scheme is indeed more effective and scalable than existing schemes.