Multi-objective optimization of integrated urban transit limited-stop pattern and bike-sharing operating strategy

Nowadays, most cities have already established well-developed transit networks and extensively distributed bike-sharing stations. To improve the efficiency of these integrated systems, this study investigates the joint optimization of transit limited-stop pattern and bike-sharing operating strategy. Considering the conflicts among users, transit operator, and bike-sharing operator, a multi-objective optimization model is formulated to optimize transit limited-stop pattern, associated operating frequencies, bike-sharing initial deployments, and dynamic relocation operations, thereby minimizing users’ costs, transit operating costs, and bike-sharing operating costs. A quasi-dynamic stochastic user equilibrium assignment model is proposed as a subproblem to formulate users’ choice behaviors under spatiotemporal heterogeneous travel demands. Additionally, an improved multi-objective evolutionary algorithm based on objective space decomposition (MOEA-OSD) is further developed to solve this complex problem. The computational results demonstrate that the proposed assignment model is capable to capture users’ choice behaviors under different combinations of operating strategies. The improved MOEA-OSD exhibits well performance in solving the multi-objective optimization problem and achieves a set of Pareto optimal solutions, providing diverse alternatives for operators to determine their final operating schemes when confronted with varying budgets and operating resources.