Intelligent Planning of Customized Intercity Ride-Hailing Service Networks

As intercity interactions become increasingly frequent, intercity travel demand continues to grow, along with the rising expectation for personalized mobility services, highlighting limitations in traditional passenger transportation modes. This study proposes a bi-level programming model for systematically optimizing intercity customized passenger transport operations. The model aims to minimize the total converted operational cost by optimizing route design, departure frequencies, and fleet-size allocations. It integrates multi-dimensional demand determinants, including demographic distribution, economic gradients, and geographic constraints, and incorporates supply-demand equilibrium constraints to intelligently capture the dynamic feedback between service supply and passenger demand. To efficiently address this complex optimization problem, the model is reformulated into a mixed-integer linear programming (MILP) formulation. A hybrid solution approach combining branch-and-cut and Frank–Wolfe algorithms is developed to effectively solve large-scale, real-world scenarios. Numerical experiments conducted in the Beijing–Tianjin–Hebei region demonstrate that the proposed method significantly improves operational efficiency and resource allocation, thereby strongly supporting smooth public mobility and sustainable regional economic development.