Transportation Scenario Planning with Graph Neural Networks

To enable data-driven scenario planning, we take the flows is, therefore, a requisite to better plan urban areas. In this first steps in leveraging the Geo-contextual Multitask Embedding context, an important task is to study hypothetical scenarios in Learner (GMEL) model, previously proposed in Liu et al. [16], as our which possible future changes are evaluated. For instance, how the base model for predicting commuting flows based on geographic increase in residential units or transportation modes in a neighborhood information (e.g., infrastructure, land use, transportation). Commuting will change the commuting flows to or from that region? In flows are defined as flows between a workers' residence this paper, we propose to leverage GMEL, a recently introduced location and a workplace location. While major cities have the resources graph neural network model, to evaluate changes in commuting to collect and process high-resolution land use data, other flows taking into account different land use and infrastructure scenarios.