[sustran] Minimum and maximum lane widths for arterials in urban areas

Tony Barry tjb at pc.jaring.my
Thu Nov 27 21:56:15 JST 1997


At 10:56 27/11/97 +0800, you wrote:
>Dear Rajeev Saraf
>
>>For Delhi, while redesigning the road cross section to accomodate bicycle
>>tracks, we had proposed 3.0m wide lanes. But there was a strong
>>resistance for few people, who said that we could not have lanes in
>>urban  areas of width less than 3.5m.
>>I would like to if lanes of 3.0m have been implemented anywhere and what
>>impact does it have on safety, speed and capacity? How have the codes in
>>various countries dealt with lane widths? Any repsonse is welcome.
>>thanks.
>>
>
>I believe you are asking about the width of the traffic lanes not the
>bicycle lanes/tracks.  I am not an engineer, so I can't answer you in great
>detail.  But I think that many cities around the world do accept urban
>traffic lanes of less than 3.5 metres.
>
snipped

>

Yes. your own home town of Kuala Lumpur for example. The Public Works
Department guide has lane widths varying from 3.5m for category U6 and U5
(full or partial separation) down to 2.75m for U2 ( minor roads - local
traffic - low volume of commercial vehicles). I would question the safety
of using low width lanes immediately adjacent to bicycle track though
(unless some physical separation is provided).

Unfortunately Malaysia has a rather high level of road accidents so one
cannot draw any conclusions as to the suitability of such lane widths from
this alone. It gets quite complex as widening lanes gives increased
opportunity for moorcylce "lane splitting" and with one of the highest
motorcycle uses in Asia here the accident rate amongst motorcyclists is
already extremely high (65% of accidents involve motorcyclists according to
a recent report).

You could try contacting Professor Karl Bang of TRE Transportation Research
and Engineering, (tre at indo.net.id) who has done extensive work on highway
capacity in Indonesia - it may contain something on effects of lane widths
but I can't be sure.





More information about the Sustran-discuss mailing list