Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Traffic Congestion Prediction Using Machine Learning Techniques

Rafed Muhammad Yasir, Moumita Asad, Naushin Nower, Mohammad Shoyaib

TL;DR

This work tackles one-week-ahead traffic congestion prediction by integrating time, day, and weather data. It employs a Support Vector Regression model with an RBF kernel trained on HERE API data for New Delhi, achieving an average RMSE of $1.12$ across four roads. The results indicate that long-horizon forecasting is feasible but hampered by limited training data, with AMWR offering stronger short-term performance. The study demonstrates the value of weather-aware features for extended horizon traffic forecasting and highlights the need for larger datasets and data fusion with real-time streams for practical deployment.

Abstract

The prediction of traffic congestion can serve a crucial role in making future decisions. Although many studies have been conducted regarding congestion, most of these could not cover all the important factors (e.g., weather conditions). We proposed a prediction model for traffic congestion that can predict congestion based on day, time and several weather data (e.g., temperature, humidity). To evaluate our model, it has been tested against the traffic data of New Delhi. With this model, congestion of a road can be predicted one week ahead with an average RMSE of 1.12. Therefore, this model can be used to take preventive measure beforehand.

Traffic Congestion Prediction Using Machine Learning Techniques

TL;DR

This work tackles one-week-ahead traffic congestion prediction by integrating time, day, and weather data. It employs a Support Vector Regression model with an RBF kernel trained on HERE API data for New Delhi, achieving an average RMSE of across four roads. The results indicate that long-horizon forecasting is feasible but hampered by limited training data, with AMWR offering stronger short-term performance. The study demonstrates the value of weather-aware features for extended horizon traffic forecasting and highlights the need for larger datasets and data fusion with real-time streams for practical deployment.

Abstract

The prediction of traffic congestion can serve a crucial role in making future decisions. Although many studies have been conducted regarding congestion, most of these could not cover all the important factors (e.g., weather conditions). We proposed a prediction model for traffic congestion that can predict congestion based on day, time and several weather data (e.g., temperature, humidity). To evaluate our model, it has been tested against the traffic data of New Delhi. With this model, congestion of a road can be predicted one week ahead with an average RMSE of 1.12. Therefore, this model can be used to take preventive measure beforehand.
Paper Structure (7 sections, 1 equation, 1 figure, 2 tables)

This paper contains 7 sections, 1 equation, 1 figure, 2 tables.

Table of Contents

  1. INTRODUCTION
  2. LITERATURE REVIEW
  3. Methodology
  4. Data Collection
  5. Development of Prediction Model
  6. EXPERIMENTAL EVALUATION
  7. CONCLUSION

Figures (1)

  • Figure 1: Comparison between Proposed Approach and AMWR