Abstract: Given the fleet size and passenger demand， a networkbased mixed integer linear programming (MILP) model was developed to simultaneously optimize the service area and route planning for the irregularshaped residential communities and the irregular road networks that were typical of many Chinese cities. The proposed model features a twolevel structure with the upperlevel to maximize the number of served passengers by the feeder transit system and the lowerlevel to minimize the operational cost for transit operators. Moreover， a heuristic approach was developed to yield acceptable solutions of the model within a reasonable amount of time. The case studies have validated the effectiveness of the proposed model as well as the algorithm， whose results reveal that the range between the nearoptimal solution by the heuristic algorithm and the optimal solution is around 15% and the difference of the operational cost for each passenger is less than 10%.

Key words: public transit, flexible feeder transit system, service area, network model