Dhaka’s Traffic Congestion: The Roads Need Eyes Too

There is little need to explain Dhaka’s traffic congestion. Every day, millions of people spend hours stuck on the roads. Office hours are wasted, students arrive late for classes, ambulances carrying patients get trapped, fuel is burned and air pollution increases. More importantly, traffic congestion has become a major burden on the economy and people’s productivity.

The question is whether Dhaka’s traffic problem is simply a matter of narrow roads. Can building more roads, flyovers and U-turns alone solve it?

We need to look at the problem differently. There is a limit to how much Dhaka’s roads can be widened. However, data-driven management could significantly improve the efficiency of the existing road network.

Having lived in London, I have seen how the city manages traffic. London also experiences congestion, but traffic management is not left solely to the police. Data is regularly used to identify when roads become busy, adjust signal timings and prioritise buses.

In its five-year plan published on January 26, 2026, London’s transport authority announced plans to expand its advanced Fusion traffic control system. By analysing large volumes of data, the system aims to adjust signals more quickly and intelligently, potentially reducing congestion-related delays by up to 14 per cent. The lesson is clear: traffic management requires real-time information from roads and decisions based on that data.

Singapore offers another important example. According to a Channel NewsAsia report published on October 31, 2025, more than 6,000 traffic violations were detected during the first month of operation of its new enforcement cameras. In 2026, Singapore’s Ministry of Home Affairs said the cameras used video analytics and automatic number plate recognition to detect offences, including running red lights and making illegal U-turns.

The key lesson is not the number of cameras but the ability of technology to monitor far more activity simultaneously than an individual police officer can. The aim is to make law enforcement more effective, not replace the police.

Malaysia is also adopting technology-based traffic management. A Malay Mail report published on May 19, 2025, described initiatives in Kuala Lumpur to analyse real-time data using AI to improve traffic flow and detect accidents and road problems quickly. Its gradual, needs-based approach is worth considering in Bangladesh.

Dubai has taken a more ambitious approach. Gulf News reported on February 25, 2025, that Dubai’s Roads and Transport Authority had introduced a traffic signal control system using AI, predictive analytics and digital twin technology. A subsequent report on September 4 highlighted significant improvements in traffic flow at selected major intersections.

Dhaka, meanwhile, has already begun moving in this direction. The Daily Star reported on April 30, 2026, that the Dhaka Metropolitan Police had introduced an AI-based system to detect traffic law violations, initially using 25 cameras in several key areas. The system was designed to identify offences such as running red lights, driving in the wrong lane, travelling against traffic and illegal parking.

On May 11, the newspaper reported that more than 300 cases had been filed within a week. A June 30 report put the number of cases recorded by the AI traffic monitoring system at around 1,500 and said more cameras were planned.

The government has also shown interest in the initiative. Prime Minister Tarique Rahman has directed officials to expedite the installation of AI cameras in Dhaka and remove old and expired vehicles. These are welcome steps. Addressing a longstanding problem through technology, modernised signals and better vehicle management deserves support.

However, installing AI cameras alone will not eliminate congestion. Dhaka’s problems also include illegal parking and road occupation, buses stopping indiscriminately, unfit and ageing vehicles, inadequate public transport, unsafe pedestrian movement and poorly coordinated traffic signals.

What the city needs now is an integrated Dhaka Traffic Command System linking key cameras, traffic signals, vehicle information and road conditions to a central platform. Real-time data could identify congestion, detect accidents and determine where signal timings need adjustment.

Ambulances and public transport could also receive priority. When an ambulance travels along a particular route, nearby signals could be temporarily coordinated to help it reach its destination faster. Prioritising buses during peak hours would allow more people to travel efficiently on the same roads.

Transparency is equally important in AI-based enforcement. Every case should be supported by photographic or video evidence, and motorists must have an accessible appeals process. Technology should serve citizens, not exercise unchecked authority over them.

Success must also be measured differently. The number of cameras installed is not the most meaningful indicator. Instead, authorities should assess reductions in journey times and signal waiting periods, improvements in ambulance response times and increases in compliance with traffic laws.

Dhaka will not become Singapore, London or Dubai overnight. Its circumstances and challenges are different. Nevertheless, lessons from these cities can help Bangladesh develop solutions suited to its own needs.

The government’s initiatives should be assessed from the perspective of public interest rather than political affiliation. If AI cameras, smart signals, controls on ageing vehicles and public transport reform become part of a coordinated plan, meaningful change is possible.

Solving Dhaka’s traffic congestion is not simply about building more roads. It is time to make road management smarter. In the cities of the future, roads will need not only more space but also eyes of their own.

The author: A politician and columnist.