Office of Operations
21st Century Operations Using 21st Century Technologies

Use of Narrow Lanes and Narrow Shoulders on Freeways: A Primer on Experiences, Current Practice, and Implementation Considerations

Appendix. Summary of Impacts Resulting from Narrow Lanes and Shoulders

Table A-1. Summary of Operational Impacts Resulting From Narrow Lanes and Shoulders.
Location Configuration Results and Lessons Learned
Houston, Texas
US 59
(Reference 1)
3- 12′ lanes to 4- 10.5′ lanes
4- 12′ lanes to 5- 10.5′ lanes
Narrowed shoulders
(Refer to Figure 1)
  • The level of service generally improved, with the major operational benefits realized in the peak periods.
  • Total delay in one section was not reduced significantly, with the bottleneck merely being shifted. (Thus the need to address the entire “system” as part of the PBPD process, and not just an isolated problem location)
  • One potential operational problem was that the altered outside lane was composed partly of the right shoulder and partly of the mainline pavement, and each of these pavement materials were a different texture and contract. At one of the entrance ramps, the contrast appeared to guide the entrance ramp traffic out of the rightmost lane.
Texas – Multiple locations in Dallas, Houston, and San Antonio, Texas
(Reference 5)
Not a before and after analysis of conversions from wider to narrower lanes; but a comparison of different roadway segments — some with 12′ lanes, others with 11′ lanes.
  • Reduction in speed of about 2.2 mph for 11-ft lanes as compared with 12-ft lanes.
  • A reduced lane and right-side lateral clearance combination directly corresponds to a reduction in the capacity of the travel lanes.
Multiple locations
(Reference 3)
NCHRP study of freeway segments throughout the US where lanes were narrowed to 11′ to increase capacity.
  • Altered sites (i.e., narrower lanes) exhibited slightly lower speeds for a given volume range and a slightly greater tendency to fall into LOS F conditions.
  • Field observations indicate that operational impacts of reduced shoulder or lane widths are most notable in the transition area (i.e., the beginning of the altered segment).
General
(Reference 17)
FHWA evaluation of the operational and safety characteristics of shoulders used as part time travel lanes.
  • The functional capacity of most shoulder lanes is approximately one-half to two-thirds that of a normal general-purpose lane, depending on the lane's geometric characteristics.
  • Speeds in the shoulder lane also tend to be 5 to 10 mi/h (8 to 16 km/h) slower than the adjacent general-purpose lanes.
  • The limited functional capacity may be due in part to the geometric deficiencies, such as narrow width, close proximity of fixed objects, or lack of continuity associated with the lane, and to the fact that motorists may feel uncomfortable using the shoulder as a temporary travel lane.


Table A-2. Summary of Safety Impacts Resulting From Narrow Lanes and Shoulders.
Location Configuration Results and Lessons Learned
Multiple locations
(Reference 3)
NCHRP study of freeway segments throughout the US where lanes were narrowed (i.e., “altered”) to 11′ to increase capacity.
  • Truck crash rates are almost always higher on altered sections (i.e., narrower lanes) compared with unaltered (12 ft. lanes).
  • Crash rates for altered sites were higher in three out of the five corridors studies, two of which used shoulders and narrow lanes on a continuous basis for an extended length (more than a mile).
  • In two locations where narrow lanes and shoulders were used to relieve specific bottlenecks and improve lane continuity, the result was a “smoothing of traffic with a better balance between supply and demand, and the crash rate decreased slightly.
  • A difference in lane width (12 to 11 feet) by itself had no significant impact.
  • Crash rates for altered sites tend to be somewhat higher than unaltered sites. However, if strategies are carefully applied in concert with lane balance and lane continuity concepts, rates for altered sections may be lower than for unaltered.
Houston, Texas
US 59
(Reference)
3- 12′ lanes to 4- 10.5′ lanes
4- 12′ lanes to 5- 10.5′ lanes
Narrowed shoulders
(Refer to Figure 1)
  • The number of crashes and the crash rates declined in the altered sections during the two years following modification for each of the four time periods studied (24 hour, peak, daytime, nighttime).
  • The larger reductions in crash frequencies occurred during the peak periods (the same period as the greatest operational benefits).
  • No significant change in the number or rate of severe accidents.
  • The upstream segment entering the modified section (with narrow lanes and shoulders) also experienced a reduction in crashes and crash rate — likely attributed to the better operations in the downstream segments where capacity had been increased by the additional lane.
  • The crash rate in the section downstream from the modified segments experienced a significant increase in the crash rate, with the greatest increase occurring the two peak hours — likely attributed to an increase in demand and flow (from the modified segments) but with no increase in capacity.
Los Angeles, California — Multiple segments
(Reference 4, 16)
5 lanes converted to 6 lanes
4 lanes converted to 5 lanes
12′ lanes to 11′ lanes
Additional lane used as HOV in nearly all cases.
(Refer to Figure 1)
  • The projects converting four lanes to five lanes, on average, resulted in increases of 10 to 11 percent in crash frequency, which was found to be statistically significant.
  • The five- to six-lane conversion projects resulted in an increase in crash frequency of 3 to 7 percent, not statistically significant.
  • The use of the added lanes as HOV lanes — and the associated increase in speed differential between the HOV and general purpose lanes — may be an explanation for the increase crash frequency.
Various locations in California, Texas, and Arizona
(Reference 19)
Lane widths reduced from 12′ to 11′ (and sometimes 10.5′) to accommodate an additional lane.
  • Analyses focused on crash rates (i.e., crashes per million vehicle miles).
  • Left shoulder removals (all but 2 feet) appear to be safe and effective capacity improvements. On severely congested freeways (ADT greater than 20,000 vehicles per lane per day) left shoulder removals appear to aid safety if congestion levels are removed.
  • The use of 11-foot lanes as aremedial measure to reduce congestion appears to operate safely.
Texas – Multiple locations in Dallas, Houston, and San Antonio, Texas
(Reference 5)
Not a before and after analysis of a conversion to narrower lanes; but a comparison of different roadway segments — some with 12′ lanes, others with 11′ lanes.
  • When comparing freeways with 12 ft. and 11 ft. lanes, there is an increase in the number of fatal and serious injury crashes in the segments with 11ft. lanes, all other roadway characteristics being equal. The increase in crash frequency ranges from 5 percent for 2-lane freeways, up to 12 percent for 5-lane freeways.
Office of Operations