Cohorting to isolate asymptomatic spreaders: An agent-based simulation study on the Mumbai Suburban Railway

• DOI: 10.48550/arxiv.2012.12839
• Publication Date: 2020-01-01
• Authors: Alok Talekar; Sharad Shriram; Nidhin Koshy Vaid...; Gaurav Aggarwal; Jiangzhuo Chen; Srini Venkatramanan; Lijing Wang; Aniruddha Adiga; Adam Sadilek; Ashish V. Tendulkar; Madhav Marathe; Rajesh Sundaresan; Milind Tambe
• Keywords: Contact tracing

The Mumbai Suburban Railways, \emph{locals}, are a key transit infrastructure of the city and is crucial for resuming normal economic activity. To reduce disease transmission, policymakers can enforce reduced crowding mandate wearing masks. \emph{Cohorting} -- forming groups travelers that always travel together, an additional policy to transmission on \textit{locals} without severe restrictions. Cohorting allows us to: ($i$) form traveler bubbles, thereby decreasing number distinct interactions over time; ($ii$) potentially quarantine entire cohort if single case detected, making contact tracing more efficient, ($iii$) target cohorts testing early detection symptomatic as well asymptomatic cases. Studying impact using compartmental models challenging because ensuing representational complexity. Agent-based provide natural way represent along with representation members larger social network. This paper describes novel multi-scale agent-based model study cohorting strategies COVID-19 dynamics in Mumbai. We achieve this by modeling urban region detailed comprising 12.4 million agents. Individual their inter-cohort they locals modeled local mean field approximations. resulting conjunction intervention simulator used assess various strategies. results quantitative trade-off between size its being. show significant benefit terms significantly impacting ridership or \&