bike advocacy, bicycle tour, bicycle safety


Planning For Bicycles
In Heterogeneous Traffic





Our content is
provided free as
a public service!

IBF is 100%
        solar powered

Follow us on Twitter


blue bar

Case Study New Dehli, India

By Geetam Tiwari and David Mozer

Street traffic in many cities, especially in less-Westernized countries, is often characterized by a diverse mix of heterogeneous vehicles (including bicycles, cargo tricycles and human and animal drawn carts). Generally all the modes find themselves in the same one, two or three lanes albeit with a near infinite variety of traffic mix, flows and densities at different locations and at different times. But there is another similarity; this traffic is characterized by lack of any effective channelization (even if lane stripes are present), mode segregation or control of speed (even if signed). In the mind of the formally trained traffic planner it is anarchy! The prognostication would be for some condition like total gridlock or total mayhem. Yet the people and goods keep getting through. And may, by some measures, actually be doing it better than in controlled conditions. And contrary to the conventional (traffic engineer) wisdom it is not anarchy.

These relatively unconstrained environments provide an opportunity to view traffic behavior in a "natural" condition (though some "socialization" may be playing a role.) Taking it one step further, by studying flow and safety characteristics of motor vehicle (MV) and non-motor vehicle (NMV) behavior on these roads we get some interesting insights into road use behavior in a situation of self-optimization.

A team from the Indian Institute of Technology, studied traffic flow and accident data for MVs and NMVs at sixteen selected mid-block sites around New Delhi. The selection criteria was designed to obtain a wide variety of traffic densities and modal mix. None of the sites had physically segregated bicycle lanes. Collection of data on speed, flow rate, densities and conflicts, was for all modes at each location. Video film was used in this analysis. Crash data was collected from the police department.

Among the observations of the study are:

  • NMVs and slow moving vehicles did not move in files and had a lot of lateral as well as forward movements. The slower the general traffic speed the greater the in-filling by NMVs. This was maximized at stops. Hence, for NMVs, flow/density measures typically applied to MVs cannot be used. To get a more accurate assessment a geometrical based measure of "aerial density" needs to be used. This definition takes into consideration not only the varying degrees of freedom of the different modes but also the varying size. Bicycles use the inside lanes only when setting up to make turns. Integration only occurs when flow rates exceed saturation rate per lane; 6000 bicycles per hour per lane, or 2000 passenger car units per lane per hour. At least at slow speeds and initial volumes, bicyclists moving into the second lane do not seem to have a negative impact on MV flows -- the bicyclists fit into the cracks -- actually increasing the volume.
  • The average speed of MVs at peak hours is much lower than the capacity speeds, but even then the bicyclists are moving at their desired speed. At peak hours the average speed of MVs doesn't vary much between high density and low density sites. MVs don't top the saturation capacity in terms of passenger-car-units-per-hour in most three lane sites but they are near capacity at some of the two and one lane sites. The bicycles at all sites are well below the saturation capacity level.
  • On two and three lane roads, as general traffic speed increases, bicycles primarily use the curb-lane-zone (CLZ) and motor vehicles do not use the CLZ even when bicycle density is low -- a defacto segregation develops. The CLZ is in fact a curb-lane-bike-zone. As the volume of bikes increases the width of curb-lane-bicycle-zone increases. Hence, the "critical mass" of bicycle use creates a defacto bike lane. The exception to this defacto segregation is buses which move into the CLZ to discharge and pick-up passengers. This creates many conflicts. (When the CLZ becomes too dense with bicycles, the buses stop in the middle of the street, which creates another set of problems as pedestrians cross the very active CLZ. Ironically, even where the engineers do not accommodate NVMs, when speeds and volumes are right, traffic behavior suggests hat segregated facilities for MV and NMV would be well received.
  • On one lane roads (3.5 m wide), the bicyclists occupy the edge of the lane -- sharing the lane with MVs. On these facilities both modes negatively impact on the maximum flow of the other mode by over 60%. Again suggesting a benefit, at least in terms of flow, for installation of a bicycle facility.
  • Analysis of accident patterns show that the self-segregation of the modes is not sufficient to ensure the safety of vulnerable bicyclists. While mid-block over-taking and side-swiping accidents are not usually a serious consideration in regimented traffic conditions, this category of accident dominates the statistics in New Delhi. Despite the proximity of MV and MNV traffic, fatalities are low, but hardly negligible. The overall statistics for bicycle fatalities in New Delhi show that 60% of bicycle fatalities occur off-peak, when traffic volumes are lower, but MV speeds are higher. 40% of the fatal bicycle accidents are during seven peak hours when volumes are significantly higher and speeds are lower (20 to 30 km/h, 12 to 18 mph). (A breakdown of peak/non-peak volume is not given.) Of the peak hour bicycle fatalities, 62% involve collisions with buses or trucks. And, of all bicycle fatalities, 73% occur at mid-block. Again this is very different from the statistic for high channelized Western cities.
  • In the situations studied, the primary justification for separating MV and NMV is safety, not capacity!

For information on the studies supporting this article, contact: Geetam Tiwari, Ph.D., Sr. Scientific Officer, Applied Systems Research Prog., Indian Inst. of Tech., Hauz Khas, New Dehli, 110016. Fax: 91-11-685-8703. Email: [email protected].  

Related article: Collision course driving in the third world, by Fred Peace, New Scientist Magazine.



Home | About Us | Contact Us | ContributionsEconomics | Education | Encouragement | Engineering | Environment | Bibliography | Essay Contest | Ibike Tours | Library | Links | Site Map | Search

The International Bicycle Fund is an independent, non-profit organization. Its primary purpose is to promote bicycle transportation. Most IBF projects and activities fall into one of  four categories: planning and engineering, safety education, economic development assistance and promoting international understanding. IBF's objective is to create a sustainable, people-friendly environment by creating opportunities of the highest practicable quality for bicycle transportation. IBF is funded by private donation. Contributions are always welcome and are U.S. tax-deductible to the extent allowed by law.

Please write if you have questions, comment, criticism, praise or additional information for us, to report bad links, or if you would like to be added to IBF's mailing list. (Also let us know how you found this site.)

"Hosted by DreamHost - earth friendly web hosting"
Created by David Mozer.
Copyright ? 1995-2020 International Bicycle Fund. All rights reserved.