We design a new transportation system where dial-a-ride services are provided in urban and rural districts with SAVs. The system aims to satisfy a city-sized passenger request set within a large road network, so an efficient dispatching algorithm is indispensable. 
We propose here a filtering mechanism to quickly identify the most worth exploring candidate SAVs to serve the request in need as well as its corresponding insertion positions. With the filtering modules combined to an adaptive best-fit insertion heuristic whose objective is to minimize the fleet size to serve all the requests, experiment results have shown that the processing time can be saved by almost 96% compared to the basic module without filters, while keeping a good dispatching quality.