The world’s leading publication for data science, AI, and ML professionals.

Impact of the Covid-19 Pandemic and the Earthquake on Traffic Flow in the Narrow City Center

On how to find hidden phenomena by mining the traffic data

Panic (Image by: Author)
Panic (Image by: Author)

Traffic data is commonly used by traffic scientists and engineers to model and predict traffic behavior. But, is it possible to extract some "hidden" phenomena that are not strictly related to traffic flow by using traffic data? In this article, the results of the paper "Mixed Impact of the Covid-19 Pandemic and the Earthquake on Traffic Flow in the Narrow City Center: A Case Study for Zagreb-Croatia"[1] are presented, where authors found interesting human behavior pattern encouraged by facing a natural disaster.

L. Tišljarić, D. Cvetek, M. Muštra, N. Jelušić: Mixed Impact of the Covid-19 Pandemic and the Earthquake on Traffic Flow in the Narrow City Center: A Case Study for Zagreb-Croatia, In proceedings of The Science and Development of Transport (ZIRP), pp. 293–300


Before we continue, give me a second of your time. If you want to support me in my work, you can do it using this secure PayPal link. Thank you 🙂

Pay Leo Tisljaric using PayPal.Me


1. Intro

2020 was a challenging year for all humanity. In my hometown (Zagreb, Croatia), alongside the COVID pandemic, a few earthquakes struck after more than 100 years without shaking. As a traffic engineer, it was interesting to observe the changes in the traffic flow in the city center, regarding the mentioned events.

As a group of researchers from the Faculty of Transport and Traffic Sciences, University of Zagreb (Leo Tišljarić, Dominik Cvetek, Mario Muštra, Niko Jelušić), we try to investigate traffic flow trends before and during the COVID-19 pandemic. Additionally, we analyze traffic flow during the earthquake, which hit Zagreb on Sunday 22nd March 2020 at 6:24 AM.

Alongside the "normal" traffic flow decrease caused by the pandemic, we found interesting human behavior pattern regarding the reaction to the earthquake.

2. Influence of the pandemic on the traffic flow

To show the influence of the pandemic on traffic flow, we present two classical traffic parameters: speed and traffic volume. For the purpose of this research, we observed one road in the city center, with the radar-based traffic counter placed here.

Figure 1. shows the speed profiles observed at the site, where every speed profile represents the average speed changing through 24h period averaged on the weekly basis (see the legend). We can see that after the 16th of March the speed of vehicles drastically increases, and the recurrent congestion in rush hours was gone. This is expected behavior due to governmental measures for restricted movement regarding the pandemic.

Figure 1. Speed profiles (Image by: [1])
Figure 1. Speed profiles (Image by: [1])

Figure 2. just confirms the expected behavior of the traffic flow. The red color shows the values averaged for 2019, while the red color shows the values during the pandemic. On the left side, we can see the relation between the traffic volume and the speed, and on the right side traffic volume profiles. Both graphs show the smaller amount of vehicles on the road for more than 50%, and therefore, higher speed values on the road.

Figure 2. [LEFT] Relations of the volume and speed for the average working day in 2019 (black) and the week with COVID-19 restrictions (red); [RIGHT] Traffic volume profiles for an average working day in 2019 (black) and the week with COVID-19 restrictions (red) (Image by: [1])
Figure 2. [LEFT] Relations of the volume and speed for the average working day in 2019 (black) and the week with COVID-19 restrictions (red); [RIGHT] Traffic volume profiles for an average working day in 2019 (black) and the week with COVID-19 restrictions (red) (Image by: [1])

3. Human behavior pattern encouraged by facing a natural disaster

After analyzing traffic Patterns during the pandemic, we analyzed the patterns during the earthquake.

Personally, I was not expecting any difference. But, I was convinced differently.

Figure 3. shows the traffic volume on the day when the earthquake stroke (red line) and the traffic volume one day before (black line). The earthquake stroke at 6:24 AM. It can be seen that traffic volume increased at 7:30 AM by more than 100%. After examining the daily press, we saw that a large amount of people just ran to their cars tried to get as far as possible. That behavior was obviously triggered by panic and fear due to the earthquake.

Figure 3. Traffic pattern that shows the increased activity on the road caused by fear and panic behavior (Image by: Authors)
Figure 3. Traffic pattern that shows the increased activity on the road caused by fear and panic behavior (Image by: Authors)

4. Conclusions

This article showed an example of how unexpected knowledge can be extracted when mining the data. We showed an example of our experiences when the human behavior pattern was extracted by mining the traffic dataset.

If you are interested in similar, traffic-related topics follow my Medium profile, or check the published research at Research Gate 🙂

If you have any questions or suggestions, feel free to comment or contact me! https://www.linkedin.com/in/leo-tisljaric-28a56b123/

Similar article:

Novel Road Traffic Anomaly Metric Based on Speed Transition Matrices

References

[1] L. Tišljarić, D. Cvetek, M. Muštra, N. Jelušić: Mixed Impact of the Covid-19 Pandemic and the Earthquake on Traffic Flow in the Narrow City Center: A Case Study for Zagreb-Croatia, In proceedings of The Science and Development of Transport (ZIRP), pp. 293–300


Related Articles