While traffic jams may seem like a thing of the past right now, stop-and-go gridlock has plagued motorists for decades and will inevitably do so again. But researchers now think they may have a solution to this problem and that lessons learned from handling COVID-19 may be to thank.
Using a physical model originally designed to analyze infectious disease outbreaks, like COVID-19, a team of physicists and engineers from across the world have applied this model to traffic patterns in six cities to better understand their ebb and flow. Using this well-established model allows the researchers to better understand the events leading up to a traffic jam as well as propose solutions that could finally make this problem a thing of the past.
"We...empirically demonstrate that traffic congestion in urban networks can be characterized using a simple contagion process."
Previous research has attempted to solve this traffic problem by using complex models and machine learning, but researchers from the new study, published Tuesday in the journal Nature Communications, say that these other models still don't capture key characteristics of the traffic data.
"Numerous studies have explored different macroscopic approaches to model the spread of traffic jams in cities, including through the lens of percolation theory machine learning methods, and queuing theory," the authors write in the study. "However, machine learning models and models based on queuing theory are unable to capture and quantitatively describe the propagation and dissipation patterns of congestion over time and space."
Instead of looking for a new model to describe the complexity of traffic patterns, the researchers instead turned to a well-known and established model: infectious disease.
"In this work, we show that traffic systems exhibit underlying spreading dynamics similar to those observed and applied in other network systems," write the authors. "Specifically, we propose and empirically demonstrate that traffic congestion in urban networks can be characterized using a simple contagion process."
The authors chose an infectious disease model called SIR (susceptible-infected-recovered) that was originally proposed by mathematicians A. G. McKendrick and W. O. Kermack in 1927 and has remained useful through the COVID-19 pandemic. The simple model works by dividing a total population into three groups -- those susceptible to a disease, those infected, and those recovered. Because the number of people infected and recovered can change over time, these three variables are created to be time-sensitive.
These three population values are then transformed into their own differential equations that scientists can use to determine their value at a given time during the course of the infection.
While traffic jams are far less serious than the diseases this model was originally designed for, the researchers say that similarities between the two still make this model a good fit. For example, in the case of traffic jams, susceptible cars are those that have not yet reached a traffic jam, infected cars are those stuck in it and recovered cars are those beyond the jam.
Using this model and applying it to traffic patterns in Melbourne, Sydney, London, Paris, Chicago, and Montreal, the researchers found that these cities still had very similar patterns of congestion despite their different geographies. Because of these similarities, the researchers write that generalizable traffic control strategies could be implemented to minimize the number of "infected cars" throughout the day.
These strategies are still being researched, but the authors write that implementing multiple different transit windows could be a solution to the gridlock and would allow for more dispersed travel times but lower overall peaks.
THE INVERSE ANALYSIS
While completely changing people's driving habits may be a hard sell -- and may be impossible for certain work schedules -- the other side of the COVID-19 pandemic may be the perfect time to test this approach. As people begin to slowly return to their physical workplaces in the months to come there may be more flexibility in people's schedules as well as a sustained caution to not overcrowd the workplace all at once. Analyzing how these changes affect traffic congestion may provide important insight into how effective these strategies really are.
Abstract: The spread of traffic jams in urban networks has long been viewed as a complex spatiotemporal phenomenon that often requires computationally intensive microscopic models for analysis purposes. In this study, we present a framework to describe the dynamics of congestion propagation and dissipation of traffic in cities using a simple contagion process, inspired by those used to model infectious disease spread in a population. We introduce two macroscopic characteristics for network traffic dynamics, namely congestion propagation rate β and congestion dissipation rate μ. We describe the dynamics of congestion spread using these new parameters embedded within a system of ordinary differential equations, similar to the well-known susceptible-infected-recovered (SIR) model. The proposed contagion-based dynamics are verified through an empirical multi-city analysis, and can be used to monitor, predict and control the fraction of congested links in the network over time.