Abstract:The study referred to in this paper focused on the problem of bus lane infrastructure design in a given road network. The main goal was to seek optimal planning (bus lane infrastructure layout) and operational strategies (such as bus frequency) in an integrated manner. The decision process involved three parties: the infrastructure planner who designed bus lane infrastructure from a public perspective, transit enterprises that operated buses on the given infrastructure from a business perspective, and travelers who determined their own routes from an individual perspective. The problem was formulated as a trilevel mathematical program, with each level corresponding to one party in the system. A heuristic solution algorithm combining a genetic algorithm and simulated annealing was developed to solve the proposed model efficiently. A case study based on a real-world road network in Beijing was implemented to test the efficiency and applicability of the proposed modeling and computing methods.