The Barapullah Phase 3 flyover was meant to improve movement between East Delhi and South Delhi, but its opening has exposed a familiar problem in Indian urban transport: adding road capacity at one point can simply move congestion to the next. Four days after the flyover opened, traffic from its six lanes was being forced into narrower carriageways, creating a bottleneck at the connection with the older Barapullah Phase 2 stretch.
The immediate problem is visible at the point where the new structure meets the existing road network. Barapullah Phase 3 has six lanes, including one lane towards Mayur Vihar and five towards South Delhi. These lanes merge into a three-lane carriageway, which then connects to the two-lane Phase 2 flyover. Vehicles from multiple lanes are therefore required to compete for limited road space within a short stretch.
The design is not only a problem of merging discipline. It is also a problem of timing. The project was originally scheduled for completion in 2017, but construction delays pushed its opening to 2026. During those nine years, vehicle numbers increased and travel patterns across East Delhi and the wider National Capital Region changed. Connections between East Delhi, Noida, Ghaziabad and South Delhi now operate within a transport system that is different from the one considered when the project was planned.
That gap between the date of planning and the date of operation is central to understanding the Barapullah bottleneck. Infrastructure is often treated as a permanent answer to congestion, but its usefulness depends on the network around it. A flyover can be structurally complete and operationally inadequate at the same time if the roads feeding into it have not evolved with demand.
Urban practitioner Hitesh Vaidya, who was previously associated with the National Institute of Urban Affairs, described the basic limitation of flyovers in simple terms: “A flyover is like shifting the traffic from the ground floor to the first floor.” The comment captures the difference between removing a conflict at one junction and reducing the total number of vehicles using the network. If the destination roads remain congested, the pressure is transferred rather than removed.
The Barapullah case also raises questions about whether the project was assessed as an independent structure or as part of a continuously changing metropolitan network. Prof P S N Rao, former director of the School of Planning and Architecture in Delhi, said a detailed survey was needed to assess the volume of traffic using the flyover and where that traffic was moving. He also questioned whether earlier surveys remained adequate after a plan made more than 20 years ago was finally brought into operation.
This is an institutional issue as much as an engineering one. A project can pass through design, procurement and construction stages while the assumptions behind its traffic model become outdated. If there is no formal mechanism to reassess demand before opening, authorities may deliver a road that reflects an earlier city. The physical structure then becomes difficult to change, even when its operational shortcomings are immediately apparent.
The experience of other cities offers a different way of thinking about road capacity. In 2006, a stretch near Birmingham in the United Kingdom addressed bottlenecks through a combination of Hard Shoulder Running and Variable Mandatory Speed Limits. Under Hard Shoulder Running, an emergency lane is opened to traffic when sensors detect congestion. Variable speed limits are adjusted according to traffic conditions to maintain gaps between vehicles and keep traffic moving.
The approach did not depend on building more lanes or adding another flyover. A Transportation Research Board assessment cited in the supplied report found that the system reduced average travel time by more than 32%, while accident rates also fell. The significance of the example is not that Delhi can directly replicate a British road system. It is that congestion management can involve operating existing capacity more intelligently, rather than treating construction as the default response.
The comparison with Seoul is even more disruptive to conventional road-building assumptions. In 2003, a six-lane highway over the Cheonggyecheon River, used by around 160,000 vehicles every day, was demolished to revive the water body. The expectation was that removing such a major road would create a traffic deadlock. According to an article published by the Behavioural Design Academy, the move was followed by a reduction in congestion and a decline of more than 45% in vehicular traffic.
That example does not establish that removing a Delhi flyover would produce the same result. It does, however, challenge the assumption that every road carrying large volumes of traffic must be expanded. Travel behaviour responds to the options available, and road capacity can influence the number and timing of journeys. The relevant policy question is therefore not simply how many vehicles a structure can carry, but how the wider transport system shapes the need to travel by car or two-wheeler.
Delhi’s current response to the Barapullah bottleneck remains operational rather than structural. Officials told the Indian Express that, based on a drone survey, the Delhi Police would manually manage the merging of lanes using barricades, traffic lights and personnel. The aim is to prevent all lanes from emptying into the narrower section at the same time. An additional traffic light is also being used.
Such measures may regulate the flow at the immediate conflict point, but they do not change the capacity of the downstream carriageway. They also place greater responsibility on traffic personnel to manage a design constraint in real time. The effectiveness of the arrangement will depend on traffic volumes, lane discipline, signal timing and the ability of officials to adapt operations as conditions change. The supplied report does not establish how long this arrangement will remain in place or whether a permanent redesign has been approved.
The deeper policy question is what Delhi should prioritise when road capacity reaches its limits. Rao argued that strengthening public transport was more useful than continuing to build flyovers. He said commuters use Barapullah because other routes, including those through Nizamuddin and ITO, are already congested. Creating more alternative routes, he argued, can attract more vehicles, while investment in additional Metro lines could address the underlying dependence on road travel.
This argument places the flyover within Delhi’s wider mobility structure. A road project is often judged by whether it reduces travel time on the corridor it serves. But if the city lacks sufficiently attractive public transport alternatives, the benefit can be absorbed by additional private vehicles or lost at the next junction. The result is a cycle in which each new road intervention creates pressure for another intervention elsewhere.
The Barapullah project’s long gestation makes that cycle more visible. A plan conceived more than two decades ago encountered delays, while the urban region around it continued to expand and its travel patterns changed. The nine-year delay beyond the original 2017 completion schedule meant that the project’s opening conditions were not the same as its planning conditions. This suggests that major transport projects need periodic reassessment not only during design but also before commissioning.
There is also a question of what counts as successful infrastructure. A completed flyover is an easily measured output. A reliable journey across the network is a more difficult outcome because it depends on several agencies, roads, junctions, public transport options and traffic-management systems working together. The Barapullah bottleneck shows how a project can achieve the first measure while falling short on the second.
For residents and commuters, the immediate consequence is uncertainty at a key connection between East Delhi and South Delhi. Traffic may be slowed at the merge, redirected through other roads or regulated by temporary measures. The supplied report does not provide traffic counts after opening or establish whether journey times have improved overall. Those figures will be necessary to determine whether the new flyover has reduced corridor-level congestion or merely relocated it.
What the evidence confirms is narrower but important: Barapullah Phase 3 has opened into a road network whose downstream capacity is smaller than the traffic volume approaching it, and authorities are using barricades, signals and personnel to manage the resulting merge. What remains unclear is whether the problem will be resolved through a permanent redesign, revised traffic operations, better demand assessment or a broader shift towards public transport. The next measure of the project will not be the length of the flyover, but whether the network beyond it allows traffic to move.