Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Graph Based Traffic Analysis and Delay Prediction

Gabriele Borg, Charlie Abela

TL;DR

This research is focused on traffic congestion in the small island of Malta, which is the most densely populated country in the EU with about 1,672 inhabitants per square kilometre (4,331 inhabitants/sq mi), and found that the DCRNN model outperforms the STGCN with the former resulting in a MAE of 3.98 and a RMSE of 7.78.

Abstract

This research is focused on traffic congestion in the small island of Malta which is the most densely populated country in the EU with about 1,672 inhabitants per square kilometre (4,331 inhabitants/sq mi). Furthermore, Malta has a rapid vehicle growth. Based on our research, the number of vehicles increased by around 11,000 in a little more than 6 months, which shows how important it is to have an accurate and comprehensive means of collecting data to tackle the issue of fluctuating traffic in Malta. In this paper, we first present the newly built comprehensive traffic dataset, called MalTra. This dataset includes realistic trips made by members of the public across the island over a period of 200 days. We then describe the methodology we adopted to generate syntactic data to complete our data set as much as possible. In our research, we consider both MalTra and the Q-Traffic dataset, which has been used in several other research studies. The statistical ARIMA model and two graph neural networks, the spatial temporal graph convolutional network (STGCN) and the diffusion convolutional recurrent network (DCRNN) were used to analyse and compare the results with existing research. From the evaluation, we found that the DCRNN model outperforms the STGCN with the former resulting in MAE of 3.98 (6.65 in the case of the latter) and a RMSE of 7.78 (against 12.73 of the latter).

Graph Based Traffic Analysis and Delay Prediction

TL;DR

This research is focused on traffic congestion in the small island of Malta, which is the most densely populated country in the EU with about 1,672 inhabitants per square kilometre (4,331 inhabitants/sq mi), and found that the DCRNN model outperforms the STGCN with the former resulting in a MAE of 3.98 and a RMSE of 7.78.

Abstract

This research is focused on traffic congestion in the small island of Malta which is the most densely populated country in the EU with about 1,672 inhabitants per square kilometre (4,331 inhabitants/sq mi). Furthermore, Malta has a rapid vehicle growth. Based on our research, the number of vehicles increased by around 11,000 in a little more than 6 months, which shows how important it is to have an accurate and comprehensive means of collecting data to tackle the issue of fluctuating traffic in Malta. In this paper, we first present the newly built comprehensive traffic dataset, called MalTra. This dataset includes realistic trips made by members of the public across the island over a period of 200 days. We then describe the methodology we adopted to generate syntactic data to complete our data set as much as possible. In our research, we consider both MalTra and the Q-Traffic dataset, which has been used in several other research studies. The statistical ARIMA model and two graph neural networks, the spatial temporal graph convolutional network (STGCN) and the diffusion convolutional recurrent network (DCRNN) were used to analyse and compare the results with existing research. From the evaluation, we found that the DCRNN model outperforms the STGCN with the former resulting in MAE of 3.98 (6.65 in the case of the latter) and a RMSE of 7.78 (against 12.73 of the latter).

Paper Structure

This paper contains 22 sections, 1 equation, 5 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Motivations and Challenges
  3. Related Work
  4. Spatio-Temporal Data
  5. Transport Malta
  6. Graph Basics
  7. GNNs Models
  8. Similar System
  9. Existing Datasets
  10. MalTra Dataset
  11. Methodology
  12. Generating Synthetic Data
  13. Data Analysis
  14. Models
  15. The Application Of The STGCN Model
  16. ...and 7 more sections

Figures (5)

  • Figure 1: Spatio-Temporal node representation over a series of time
  • Figure 2: Data points of one user (light blue points) extracted from Google Timeline - an outline of Malta can be seen (including Saint Julian's, Valletta and the Three Cities) - own source
  • Figure 3: Generating new data points between the existing ones, referred to as 'checkpoints', which are obtained from Google Timeline.- own source
  • Figure 6: Difference in traffic duration by three-time interval groups (Floriana) - own source
  • Figure 7: Marsa northbound traffic flow comparison between MalTra and Figure 4-19 in research by author Nigel Pace pace2017investigating