The controversy over Bengaluru’s Hebbal tunnel study is no longer only a disagreement over one road project. It has exposed a deeper problem in how large transport interventions are evaluated: the authority commissioning the assessment, the consultant preparing it, the data used in the model and the surrounding projects must all fit into one credible decision-making framework. In the case of the proposed Hebbal short tunnel and the Mehkri Circle elevated corridor, Indian Institute of Science (IISc) experts and city mobility specialists say that framework remains incomplete.
The Bangalore Development Authority (BDA) commissioned an August 2026 study led by IIT-Bombay professor Archak Mittal after the Hebbal short tunnel attracted public opposition. The report, titled Integrated Traffic Micro-simulation Study – Bengaluru: Hebbal Flyover & Hebbal Short Tunnel, was prepared for Rithwik Projects Pvt Ltd. According to the Times of India report, the study concluded that the tunnel and elevated corridor were required for future traffic. A BDA official described it as a positive third-party institutional evaluation and said the project is expected to be completed by the end of January 2028.
But the study’s institutional association has not settled the debate. Prof Ashish Verma, convener of the IISc Sustainable Transportation Lab, reviewed the report and identified gaps in its traffic data, methodology and assessment of benefits. Former chief technical adviser to the Directorate of Urban Land Transport, Satyajit Arikutharam, described the study as being of extremely poor quality and raised questions about both the assessment and the detailed project reports on which it relied.
The dispute matters because a traffic model does not simply describe congestion. It helps determine whether a project is needed, how it should be designed and what consequences it may create elsewhere. If the input data is outdated, if important routes are not represented, or if only a limited number of scenarios are tested, the model can produce a technically precise answer to an incorrectly framed question.
One of the clearest concerns relates to the speed assumptions in the study. Verma’s review says the model reported peak-hour traffic speed in the 2027 base case at 13.7 kmph, while the DPR for the Hebbal short tunnel reported 4.3 kmph. The two figures represent a substantial difference in the starting picture of congestion. The supplied report does not establish which figure more accurately represents present conditions, but the gap itself makes the model’s baseline a central issue in judging the project’s benefits.
The source of the data is also significant. According to Verma’s review, the traffic information used in the IIT-Bombay study came from DPRs provided by the contractor. Arikutharam said fresh survey data was necessary because the traffic situation had changed after an additional ramp at the Hebbal flyover opened in January 2026. The IIT-Bombay study was conducted in August, but the project assessment therefore appears to have relied substantially on data that may predate a major change in the road network.
That timing raises a basic institutional question: when a project is being reconsidered after public criticism, should its assessment use the most recent available conditions rather than reproduce the assumptions of the original DPR? The report does not provide evidence of a new origin-destination survey. Arikutharam said neither DPR contained such a survey, even though knowing where travellers begin and end their journeys is essential to understanding whether traffic would use the tunnel, the elevated corridor, the existing underpass or surrounding roads.
Without that information, the model can estimate movement through a network but may not adequately explain the reasons for that movement. This distinction is important at Hebbal, where different streams of traffic have different destinations and where the usefulness of a new link depends on how directly it serves those journeys. A project may improve movement for one turning movement while leaving through traffic on the existing road system, or it may shift queues rather than remove them.
The proposed geometry described by Verma illustrates this problem. He said there is no ramp on the proposed elevated corridor for vehicles travelling towards Windsor Manor. As a result, only vehicles turning from Mehkri Circle would be able to use the flyover, while a substantial share of vehicles travelling straight would continue to depend on the existing underpass. This is not a question that can be answered by looking only at the project’s total capacity. It requires an assessment of individual movements, destinations and the points at which congestion is likely to reappear.
The scenario design is another weakness identified by Arikutharam. He said the study was expected to examine four situations: the existing base case, each of the two projects separately and both projects together. Instead, he said, it assessed only the base case and the combined tunnel-and-elevated-corridor scenario. Testing each intervention separately is important because it shows whether one project is doing most of the work, whether the second project adds measurable value or whether the combined scheme produces benefits that cannot be attributed clearly to either component.
Queue lengths were reportedly not presented before and after the proposed interventions. Speed is only one measure of network performance. A road can show an improved average speed while queues become longer at a downstream junction, a merge or an unserved approach. The absence of queue-length information makes it harder to assess whether the proposal reduces congestion across the corridor or shifts it to adjacent roads.
The study’s treatment of vehicle categories has also been questioned. Arikutharam said the DPRs and the IIT-Bombay study used passenger car unit conversion factors dating from the 1990s. He argued that these assign higher conversion values to two-wheelers than newer Indian Roads Congress guidelines recommend, potentially inflating their estimated contribution to congestion by around 30%. The supplied report does not independently verify that calculation, but it identifies the conversion factors as a material modelling choice rather than a minor technical detail.
Such assumptions matter particularly in Bengaluru, where two-wheelers form a large part of daily traffic. The way motorcycles are converted into a common traffic unit affects the apparent demand on road space and can influence the scale and type of intervention considered necessary. It also shows why a traffic study must disclose its inputs clearly enough for external reviewers to reproduce or challenge the result.
The criticism extends beyond the two projects themselves. Verma questioned how the Mehkri Circle elevated corridor would integrate with the larger east-west corridor proposed between Yeshwanthpur and KR Pura. Arikutharam said the IIT-Bombay study did not consider the impact of a Metro line and the proposed long tunnel in its traffic projections. If related projects are left outside the model, the resulting demand estimate may describe a network that will not exist when the Hebbal works are operational.
This is the larger planning issue at the centre of the dispute. Bengaluru’s transport decisions are distributed across multiple institutions and project packages. A BDA road project can affect roads, flyovers, Metro access and future corridors managed or proposed through other agencies. Evaluating each project independently may be administratively convenient, but traffic does not recognise institutional boundaries. It responds to the combined network available to motorists.
The BDA’s decision to commission a further study after criticism is therefore significant, but a third-party label alone cannot establish independence or adequacy. The credibility of the exercise depends on the freshness of the data, transparency of the methodology, completeness of the scenarios and disclosure of how the findings relate to other planned interventions. The reported recommendation to increase the distance between the elevated corridor and the tunnel from 80 metres to 250 metres also suggests that the assessment could require changes to the existing design.
For the public, the central issue is not whether the Hebbal tunnel has institutional support or whether congestion is a serious problem. It is whether the proposed response has been tested against current traffic conditions and the full set of alternatives and related projects. The BDA says the works are necessary for future traffic and are scheduled for completion by January 2028. The IISc review and mobility experts, however, have identified unresolved questions about the evidence used to reach that conclusion.
The available material confirms that the Hebbal project has received a new traffic assessment and that the assessment is being contested on specific technical grounds. It does not establish whether the proposed tunnel and elevated corridor will ultimately improve the wider network, because the competing claims have not been settled through the evidence provided. The milestones to watch are the project’s design response to the study, the treatment of the proposed 250-metre separation, the availability of updated traffic and origin-destination data, and whether Metro and other planned corridors are incorporated into the final decision.
