COVID-19 disproportionately affects the low-wage workers whose spatiotemporal mobility pattens, e.g., between housing and job, have dramatically changed. Microtransit service has been recently launched in Detroit to complement the existing public transit options. Despite the initial success, a salient issue is how to effectively and efficiently utilize microtransit resources to ameliorate spatiotemporal mismatch between employment and housing for low-wage workers. With the rise of Artificial Intelligence (AI) and increasingly available smart mobility data, the vision of this research project is to create a dynamic routing prediction system based on learning the hourly mobility patterns between jobs and housing. It is designed for the stakeholders (i.e., community advocates and public transport authority) to visualize and forecast the mismatch between employment and housing, which is translated into a dynamic trip demand that can be used to design adaptive routing algorithms to optimize the allocation of microtransit resources and to enhance micromobility via minimizing the rider’s first/last mile.
Currently public transportation with fixed routes and schedules are periodically tweaked and/or augmented to ameliorate the ever-changing spatial mismatch. Despite its long-term effectiveness, it is not sufficiently flexible to adapt to the hourly spatiotemporal variation of jobs-housing mobility patterns primarily from the hourly paid workers. The long-term goal of this project is to work with civic partners in the city of Detroit to (1) design, implement and deploy an AI-assist microtransit system to ameliorate the spatiotemporal mismatch between housing and employment, particularly for the low-wage workers residing in the under resourced neighborhoods; and (2) use geocoded socioeconomic data to identify the community with disparities in mobility and deploy smart mobility technology to reduce the disparities and foster thriving communities. The project’s near-term objective is to leverage and power the existing microtransit service with cutting-edge technology and select a few spatiotemporally mismatched regions in Detroit as the testbed for our smart mobility strategy.
The research innovation is expected to provide immediate, low-cost yet effective public transit solutions that are expected to bring an immediate benefit to the vulnerable communities in Detroit by significantly reducing transit risk, commute time/distance and trip cost. It can be replicated to other US cities to ameliorate the spatiotemporal mismatch between housing and employment. In addition, it can provide insight for designing long-term intervention strategies to eliminate the mismatch and reduce the mobility disparities, for example, government to launch new transportation options and create jobs; and builders to develop housing in the mismatched regions.
This project is in response to Track A – CIVIC Innovation Challenge - Communities and Mobility a collaboration with NSF and the Department of Energy.
This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.
