# Predictability of road traffic and congestion in urban areas

**Authors:** Jingyuan Wang, Yu Mao, Jing Li, Chao Li, Zhang Xiong, Wen-Xu Wang  
**arXiv:** [1407.1871](https://arxiv.org/abs/1407.1871)  
**Date:** 2014-07-04  
**Summary page:** https://www.summarizepaper.com/en/paper/1407.1871v1/

## Key points

- Mitigating traffic congestion in urban areas is crucial for energy consumption reduction and air pollution control.
- Predicting road usage and traffic conditions based on drivers' collective behavior is key to addressing congestion issues.
- Study by Jingyuan Wang, Yu Mao, Jing Li, Chao Li, Zhang Xiong, and Wen-Xu Wang explores predictability of road traffic in urban areas using precise daily vehicle mobility records.
- Mapping congestion levels onto a time series of symbols and measuring entropy revealed high daily predictability in traffic conditions.
- Road segments with intermediate average travel speeds were found to be the most challenging to predict.
- Possibility of predicting traffic condition of inaccessible segments based on adjacent segment observations was investigated.
- Urban traffic patterns are highly predictable despite diverse driver behaviors and destination variability.
- Research contributes valuable insights into developing accurate predictive models for mitigating urban road congestion effectively.

## Summary

In urban areas, mitigating traffic congestion on roads is crucial for , energy consumption reduction, and air pollution control. The ability to predict road usage and traffic conditions based on drivers' collective behavior is key to addressing this issue. A study by Jingyuan Wang, Yu Mao, Jing Li, Chao Li, Zhang Xiong, and Wen-Xu Wang explores the  of road traffic in urban areas using precise records of daily vehicle mobility from . By mapping congestion levels onto a time series of symbols and measuring entropy, the researchers found a high level of daily predictability in traffic conditions. This predictability was observed despite the lack of prior knowledge about drivers' origins and destinations and the varying travel patterns between weekdays and weekends. Interestingly, the study revealed that road segments with intermediate average travel speeds were the most challenging to predict. Additionally, the researchers investigated the possibility of predicting the traffic condition of inaccessible segments based on adjacent segment observations with limited visibility. The findings suggest that even with diverse driver behaviors and destination variability, urban traffic patterns are highly predictable. The results have significant implications for developing accurate predictive models to devise practical strategies for mitigating urban road congestion effectively. Ultimately, this research contributes valuable insights into understanding and managing urban traffic dynamics for .