Thane: Adaptive Traffic Signals for Managing Traffic

December 30, 2020

Author: Information_Technology

Problem/Challenge

Excessive Traffic congestion issues were observed at various intersections across Thane City due to predefined/fixed signal time irrespective of traffic intensity.

Managing Traffic and Transport through Crowd-Sourced Data

Based on the need presented in the above section, the core objectives set for the project “Managing Traffic and Transport through Crowd-Sourced Data” are as follows:

• Comprehensive data collection of traffic characteristics such as speed and travel times through video cameras
• Data collection through crowdsourced eg. Google data
• Validation of crowdsourced speed and travel time data with fixed infrastructure
• Using crowdsourced data for real-time traffic signal optimisation

Methodology:

Figure 1: Methodology for Solution Implementation

 

Google Data Collection:

Google aggregates speed data from smartphone users to estimate speed and travel times. Figure below shows the representation of the Google information flow.

Figure 2: Representation of Data Extraction Process

This data is processed into traffic data using different Application Programming Interface (API), developed by Google, such as distance matrix, directions, speed, and many more. These APIs provide functionality like data analysis, machine learning services (Prediction API) or accessing user data (when permission to read the data is given). The flowchart presented in Figure 2 explains the data extraction process.

The Google API calculates a representative speed value from the available crowd-sourced data on a link at any time of the day. Each link is identified by a unique string referred to as the place ID. A road section is broken down into multiple place IDs at the locations where the road geometry or homogeneity undergoes a change (for example, merge and diverge points, intersections).

Custom-made controllers were installed at these intersections and different strategies including real-time adaptive and synchronization were tested.

Insights obtained from Data

Adaptive Traffic Signals:

Traffic data collection both before and after installation of controllers is critical for performance assessment of crowdsourced adaptive traffic signals. The project team placed video cameras at both Almeda and Khopat junctions to extract information on traffic volume (number of vehicles passing through the intersections). Traffic videos were collected for one week before installation and also during the trial period. Also, personnel were placed at the intersections to note down the queue length at all the approaches.

Figure 3: Queue Length Reduction

Following insights were received from the analysis of traffic data collected at various junctions:

Figure 4: Traffic Volume Improvement

Figure 5: Delay Reduction

Figure 6: Fuel Cost Savings

 

Figure 7: Travel Time Savings

Conclusion:

Hence, the analysis of data from various sources including adaptive traffic signals, CCTVs, Google APIs has led to multiple benefits for the city like Queue Length Reduction, Traffic volume Improvement, Delay Reduction, Fuel Cost Savings and Travel Time Savings.