Bus Rapid Transit (BRT) system is a kind of Intelligent Transportation System (ITS), which combines wireless communications, computational technologies, floating car data/floating cellular data technologies, inductive loop detection and so on. The GPS (Global Position System) data and the loop detection data are arriving at the BRT dispatch center massively at the same time so the dispatcher can know the real-time status of buses. The passenger flow data are also captured by devices deployed in the stations and then be passed to center too. How to handle those data at low latency and get some events to be used in bus dispatching is really a problem. In this paper, we propose a framework using ESP (event stream processing) technologies that solves the problem more effectively. At last we demonstrate a prototype system implementation that applies in BRT Line 1 project in Beijing.