Identification and Analysis of Weather-Sensitive Roads Based on Smartphone Sensor Data: A Case Study in Jakarta

Area of Science:

• Urban planning
• Transportation engineering
• Environmental science

Background:

• Weather-induced traffic congestion is a significant issue in metropolises, particularly those with inadequate drainage.
• Limited research exists on weather-sensitive roads (WSR) in developing countries with potentially deficient infrastructure.
• Identifying WSR is vital for targeted infrastructure development and traffic management.

Purpose of the Study:

• To identify weather-sensitive roads (WSR) in Jakarta, Indonesia, using real-world data.
• To analyze the correlation between weather changes and traffic congestion patterns.
• To understand the contributing factors to WSR in a developing country context.

Main Methods:

• Utilized smartphone sensor data, real-time weather information, and road characteristics.
• Developed a spatial-temporal congestion speed matrix (STC) to visualize traffic speed fluctuations.
• Applied a sequential clustering and classification framework to identify WSR based on traffic speed patterns.

Main Results:

• The proposed STC and classification framework effectively identified WSR by clustering roads with similar traffic speed changes due to weather.
• Key factors highly correlated with WSR in Jakarta were identified.
• Found significant correlations between WSR and proximity to shopping malls, mosques, schools, as well as road altitude, length, width, and number of lanes.

Conclusions:

• The study successfully identified weather-sensitive roads in Jakarta, providing valuable insights for urban planning.
• Road geometry and proximity to key urban features are critical determinants of weather-induced traffic congestion.
• The findings offer a data-driven approach to enhance urban infrastructure development and mitigate traffic issues in developing cities.