Shared Mobility Optimizer

With the transportation sector currently the leading source of emissions among major industries, shared mobility has a crucial role to play in reducing emissions and ensuring energy efficient, affordable, and equitable mobility for all. But implementing shared mobility effectively and at scale faces significant barriers: inconsistent data, complex network interactions, the absence of consistent frameworks for assessing project outcomes, and the difficulty of assessing emerging technologies amid rapid progress in automation, electrification, and connectivity. The Shared Mobility Optimizer (SMO) harnesses state-of-the-art machine learning, graph analysis, predictive algorithms, and data management techniques to provide communities with an accessible and cost-effective web-based tool capable of evaluating and optimizing a wide range of shared mobility technologies and projects in the context of the entire transportation system. SMOĺs mission is to revolutionize transportation planning with a comprehensive tool that not only optimizes shared mobility projects for efficiency and connectivity, but accelerates the decarbonization of local transportation, ensuring equitable access and maximizing the impact of limited resources. The final developed technology leverages modern machine learning models that are capable of predicting the mode share of cars, transit, micromobility, microtransit, and even active transportation, within the context of the local transportation system, helping planners prioritize, win funding, and optimize their mobility projects. Transportation planners are able to interact with these models through an intuitive web interface that allows non-technical users to complete robust analysis of projects in their community. The Phase 1 research created a data management system, model training pipelines, and algorithms for impact analysis, optimization, and project generation. We validated the accuracy of models with real world aggregate mode share data as well as with regional travel demand models. Finally, we validated the commercial demand for this technology through pilot projects and stakeholder outreach. In Phase 2 of the project, (i) analysis will be expanded to improve modeling accuracy, support new modes, and identify shared mobility gaps, (ii) a scalable model governance framework will be created ready for production deployment, (iii) a market-ready web application will be developed to facilitate the accessible and intuitive analysis of shared mobility projects, and (iv) the product will be commercialized and demonstrated through pilot projects and partnerships. The two pilot projects completed during Phase 1 demonstrate how SMO has the capacity to make shared mobility project planning faster, cheaper, and more comprehensive. Improving the planning process has significant downstream effects, leading to more shared mobility projects and higher adoption, decarbonizing transportation and bringing affordable, clean mobility to communities across the US.