Spatiotemporal Modeling of Disease Spread Through Micromobility Systems
Analytics
23 views ◎10 downloads ⇓
Abstract
New modes of public transportation such as micromobility are rapidly growing in urban areas. Bike sharing and e-scooter sharing, for example, have been advanced to solve the first/last mile problem, providing quick access to bus stops and train stations for their users. This efficiency, however, may come at the cost of transmitting disease since the surfaces on the bicycles or scooters are subject to germs and harmful pathogens when they are left in contaminated places or used by infectious individuals. This dissertation aims to understand various facets of the role of micromobility transportation in the spread of viral disease within dense urban areas. I propose a novel micro-level and spatially-explicit agent-based modeling framework to model the spread of viral infectious diseases through micromobility systems and a baseline population. I use this simulation framework to study the role of micromobility in the spread of viral disease in urban areas by breaking down the problem into three directions. First, I want to study how surfaces on the new micromobility transportation systems contribute to the emergence and dynamics of viral epidemics in urban areas. Second, I seek to find out how geographic space and time are organized concerning the risk of exposure to a viral disease out of using micromobility vehicles. Third, to inform decision-making in response to the spread of viral disease through micromobility systems, I examine what intervention methods and strategies, including random or systematic intervention, are more effective in controlling the spread of infectious diseases through micromobility vehicles. In order to test the proposed model, a case study is conducted in Cook County, Illinois, and uses the Chicago City public bikesharing system. Results show that the emergence of viral disease through micromobility transportation in Cook County is possible, but the overall impact of the system on the disease dynamics in a worst-case scenario, especially with the current size of the system, is rather small. The proposed model, however, provides a better measure to evaluate the role of transportation in spread of disease compared to existing measures. The spatial pattern for the risk of exposure is higher in the central business district and in northern regions, where most of the shared bike transportation occurs. Moreover, the start day of exposure impacts the dynamics of the spread of disease through both micromobility and the baseline population. Finally, intervention success in a full-blown epidemic highly depends on human behavior, availability of disinfection equipment, and strategies to implement control methods. The proposed simulation framework can be used to assess the efficacy of interventions and make trade-offs between these factors when dealing with epidemics of the sort analyzed in this research.