AI and Traffic Management: A Call for Innovation Inspired by Our Youth

We have a compelling story to tell the world, a story comes back again and again in August— an incredible courage and spirit of our youth. In the month of August 2019, ordinary students from different parts of the city took to the streets following the tragic death of two students of Shaheed Ramiz Uddin Cantonment College in Dhaka. They called for a ‘safe road movement’ demonstrating an innovative traffic management system. A beautiful trophic system is shown by young students that caught everyone’s attention. Remarkably, we witnessed a repeat of the exact same incident on August 2024. After the anti-discrimination student movement and the fall of the government, when the normal activities of the traffic police broke down, the students came forward to restore traffic order. They operated in a more orderly and efficient way to control the traffic at every intersection. They managed several activities including lane management, proper use of left lane, ensuring helmet wearing, driving license checking and so on. Now this rises a common question is when will we implement the practical and long term-traffic management shown by the youth?

Here we mean traffic management by standing traffic police on the roads and activating the intersections. This traditional approach is causing long traffic jams on one hand and putting the traffic police personnel at long term health risk on the other. According to a research report, 84.4 percent of traffic policemen are suffering from respiratory problems due to exposure to pollution, and many are at risk of injury or death due to unsafe conditions. Moreover, the traffic system controlled by manual observation without any advanced technical system exacerbates additional traffic congestion. Usually, police officers cannot decide which lanes to keep open for how long because there is no precise information about the extent of traffic congestion. According to a survey by the BRAC Institute of Government and Development estimated that traffic congestion costs 5 million work hours per day with an economic loss of about $11.4 billion annually. So what can be the way out of the problem?

A possible answer lies in artificial intelligence (AI) and strategic policy reforms. All these problems and an orderly traffic system showing youth can be implemented through the application of artificial intelligence (AI) and implementation of some policy decisions. The application of artificial intelligence (AI) in the implementation of traffic regulations has been in vogue in different countries worldwide for a long time. We are far behind in that regard. Even our current traffic management system is outdated and lacks technological integration. In such a situation, an automatic traffic management directly through AI technology might seem like a dream, but it is not unattainable. It needs some initiatives and policy reforms. Where a strong gateway of seamless flow of information and cooperative attitude of concerned bodies will be ensured.

Our primary challenge in traffic management is getting drivers to obey the law. Problems such as illegal parking, improper lane usage, non-functional traffic lights, and the inability to penalize violators are common.  These problems could be easily solved with the help of artificial intelligence (AI) and computer vision technology. With the help of computer vision technology, it is possible to recognize the information of each vehicle and take decisions through real time video analytics on the feeds of street CCTV cameras. AI can monitor traffic conditions and detect violations, functioning like artificial traffic police and automatically issuing fines to offenders.

Let’s take a look at how it works. Suppose there are surveillance cameras on the city streets. With which the overall picture of the road can be captured. Think of each camera as the eyes of a traffic policeman. A traffic policeman after visually capturing the law-abiding vehicles makes a judgmental decision and identifies them. For that traffic police need intelligence / neuron and traffic related knowledge / logic. Likewise, when you add analytics servers and deep learning algorithms to surveillance cameras, you start making intelligent decisions. Let’s know what benefits we can enjoy with the help of artificial intelligence.

Adaptive Traffic Control System: Automatically analyzes the extent of congestion on each road and optimizes the traffic signal to minimize congestion and save commuters from wasting extra time at a signal.

Intelligent Route Direction: Drivers will be able to analyze the current vehicle speed on the road and the flow of signals to provide relatively easy and less congested route detection, which plays a role in reducing excess traffic on a particular road.

Predict Traffic Congestion: Analyzing historical data can also be an effective measure to reduce traffic congestion by alerting citizens about the most congested times and places in the city.

Vehicle Identification: Any vehicle can be identified and movement monitored through automated number plate recognition. It helps to locate any suspicious vehicle or vehicle involved in organized crime at the earliest.

Automated Law Enforcement: As a vehicle can be detected automatically with the help of artificial intelligence, it is possible to file an automated case against a specific vehicle if it violates traffic laws such as lane violation, speeding or signal violation.

Accident Detection: Fully automatic detection of any type of accident and immediate action to be taken by alarming the control room.

Implementing these AI-driven solutions could significantly improve traffic management and help build a modern, livable city. However, several challenges remain.

A significant issue in our traffic management is the lack of inter-departmental coordination. For instance, while the city corporation is responsible for installing traffic lights, the police controls the traffic. Even if the police control the traffic, the Bangladesh Road Transport Authority (BRTA) holds vehicle information. Meanwhile, BRTA has vehicle information, citizen information is managed by the Election Commission. In this situation, the development of traffic management cannot be imagined without the continuous coordination of these institutions. . Additionally, for AI systems to function correctly, number plates must be clearly visible, which may necessitate redesigns and a central database that integrates BRTA and National ID data. Furthermore, consolidating all traffic-related responsibilities under a single organization could streamline efforts and improve efficiency.

The youth of Bangladesh have shown us what is possible with determination and innovation. Now, it is time to take their lead and leverage AI to create a safer, more efficient, and economically vibrant city for all.

Md. Mohiuddin Shiblee
Researcher (Smart Policing, AI Analytics, Intelligent Security)
Chief Business Development Officer at Trimatric AI