Schedule-based Analysis of Transmission Risk in Public Transportation Systems

TL;DR

This paper presents a transmission risk modeling framework for public transportation using the Wells-Riley model, analyzing how operational and external factors influence airborne disease spread, with case study insights from the MBTA Red Line.

Contribution

It introduces a quantitative, adaptable model for assessing transmission risk in transit systems based on operational data and virus characteristics, aiding policy decisions.

Findings

Increasing frequency reduces transmission risk but cannot offset high infection rates.

Uneven passenger distribution increases overall infection probability.

Demand balancing among train branches can lower transmission risk.

Abstract

Airborne diseases, including COVID-19, raise the question of transmission risk in public transportation systems. However, quantitative analysis of the effectiveness of transmission risk mitigation methods in public transportation is lacking. The paper develops a transmission risk modeling framework based on the Wells-Riley model using as inputs transit operating characteristics, schedule, Origin-Destination (OD) demand, and virus characteristics. The model is sensitive to various factors that operators can control, as well as external factors that may be subject of broader policy decisions (e.g. mask wearing). The model is utilized to assess transmission risk as a function of OD flows, planned operations, and factors such as mask-wearing, ventilation, and infection rates. Using actual data from the Massachusetts Bay Transportation Authority (MBTA) Red Line, the paper explores the transmission risk under different infection rate scenarios, both in magnitude and spatial characteristics. The paper assesses the combined impact from viral load related factors and passenger load factors. Increasing frequency can mitigate transmission risk, but cannot fully compensate for increases in infection rates. Imbalanced passenger distribution on different cars of a train is shown to increase the overall system-wide infection probability. Spatial infection rate patterns should also be taken into account during policymaking as it is shown to impact transmission risk. For lines with branches, demand distribution among the branches is important and headway allocation adjustment among branches to balance the load on trains to different branches can help reduce risk.