Efficient transit planning framework using service composition and hierarchical aggregation schemes

Modern advanced public transportation systems (APTSs) mostly maintain their own regional service information in a closed, non-sharable fashion in order to provide path planning service. When processing large-scope cross-regional paths which would require data from two or more APTSs, the service would be difficult or impossible since there is scarcely protocol for cross-system data exchanging. In addition, when planning a large scope trip, huge amount of services would take a long time to process. This work proposes NimbleTransit, an efficient web service-based transit planning framework using dynamic service composition and hierarchical aggregation schemes. In the framework, web service interfaces are specified for information exchanging, making it possible to integrally collect heterogeneous data from various APTSs and transportation enterprises. A dynamic service composition scheme is designed where the transit planning is performed between semantic and syntactic level to achieve scalability. The system prototype is implemented using web service-based data from Institute of Transportation in Taiwan, and the results showed that the proposed scheme is indeed more effective and scalable than other schemes.