Large-Scale Multimodal Transit Package Optimization : The Case of Hong Kong

Mobility-as-a-Service (MaaS) that allows travelers to access multiple mobility services through an integrated platform has been regarded as a promising solution to building seamless multimodal transport systems. Current MaaS business models mainly provide integrated mobility packages to travelers, then allocate the aggregate revenues to participating mobility service operators. These business models are only financially viable when participating service operators can be better off and the package operator can achieve its operation goal. To evaluate the financial feasibility and the society impacts of mobility packages, we develop a comprehensive activity-based traffic model for the large-scale mobility package optimization problem with a specific focus on transit packages. We incorporate demand elasticity in the modeling framework and capture the changes in travel demand, transport system travel time, social welfare and transit service operators’ profitability before and after the deployment of transit mobility packages. A gradient-based optimization framework using the iterative backpropagation technique is developed to solve the high-dimensional optimization problem efficiently. We perform numerical experiments using real data on the large-scale multimodal network of Hong Kong Island and analyze two package instances with government and commercial operator as the package operator, respectively. The results indicate that transit package can reduce total system travel time and increase social welfare in both instances. We also find that transit package can induce travelers to reduce auto trips and increase transit trips