One of the main elements of a road cross section are the traffic lanes – indeed, the traffic lanes are the only reason the road exists. After a century or so of developing roads you would expect there to be some sound practical concept for deciding what the width of these traffic lanes should be.
Three approaches could be considered:
• Lane width based on figures from history
• Existing standards and guidelines
• Widths developed from first principles
This post comments on “historical practice”.
If we look at historical practice we could start with the Roman empire, around 2,000 years ago. Lay (ref.771 page 13) wrote that:
“Neolithic Cretan and Roman roads indicate a wheel guage of 1.4m, which is consistent with a vehicle width of a little under 2m”
Another source (Ref. 260, quoting Mumford, 1991) said that:
“The width of Roman roads varied from 8 feet (2.5 m) to 24 feet (7.5 m) wide on parts of the great trunk highways, but generally the standard width was 15 feet (5 m)”
Today a Renault Twingo has a (front) track of 1.4m and a basic width of 1.7m, whilst one estimate of the average width of a sample of 132 modern cars is 1.8m (own research), whilst two lane urban roads might be as low as 4.1m (ref. 60).
In some notes on road geometry 1900 – 1940, Wolhuter (ref. 1046) wrote that the UK’s Ministry of Transport recommended a lane width of 10 ft. In 1973, the UK’s “roads in urban areas” (ref. 61) recommended lane widths of
- 9 ft (2.75m)
- 10 ft (3,05m)
- 11 ft (3.35m)
- 12 ft (3,66m)
The 2011 edition of the AASHTO green book (ref. 831) says:
“Lane widths of 2.7 to 3.6 m [9 to 12 ft] are generally used, with a 3.6-m [12-ft] lane predominant on most high-speed, high-volume highways”
The same widths (and virtually the same wording) were used in the 1994 edition of the green book , and Hauer (ref. 765) seems to suggest that the same widths were also referred to in the 1954 edition.
Looking at the figures for Roman roads, maybe there has not been much of a change in vehicle sizes over the last two thousand years, so that there is no need to expect a big change in lane widths (and there doesn’t seem to have been a big change either).
As if to support this idea, in 2011 the AASHTO green book was still quoting the same lane widths as appeared in the UK some 70 or so years earlier. Even in 2013, one new manual on highway geometric design which appeared in 2013, prepared with the help of international consultants, still referred to a 3.65m (12 ft) lane width (ref.1505).
It looks like either the developers of standards show a certain inertia in the selection of lane widths, or there are reasons for consistency over time. Perhaps this is partly due to vehicles being designed to fit the roads, rather than the roads being modified to fit new designs of vehicles.
(Note: so far we have only been talking about straight traffic lanes).
60 – UK, DB 32 Residential roads and footpaths, HMSO; 1978
61 – UK, “Roads in urban areas (3rd impression)” HMSO; 1973
260 – taken from an Internet document “the street in history”, author and other details not known
711 – M.G. Lay, “Handbook of road technology vol.1 planning and pavements (2nd edition)”, Gordon and Breach; 1990
765 – Canada, E. Hauer, “Safety in geometric design standards”; 1999
831 – USA, “A policy on the geometric design of highways and streets” AASHTO; 2011
1046 – South Africa, K. Wolhuter, “Wolhuter on Geometric Design”, article in Civil Engineer magazine; 2013
1505 – Nigeria, “Highway manual part 1 Design / vol. I: geometric design”, Federal Ministry of Works; 2013