Abstract:Geometric configurations of highways such as cross-sectionalelements affect traffic characteristics such as flow, density, and speed ofvehicles. This review paper is an attempt to understand the effects of lanewidth on speed of vehicles. Several papers are mentioned and they illustrates thatlane width of highways affect the operational characteristics of highways suchas speed, capacity, etc. 1. Introduction The lane width of a roadway affects the comfort of driving,operational characteristics, Level of Service (LOS) and in some circumstances,the likelihood of crashes.
Cross-sectional elements of highways consistof: travel lanes, shoulders, cross slope,barriers, curbs, gutters, sidewalks, and medians. A travel lane should have suitable width that isable provide traffic demand. Table 1 provides the values forthe width of traveled lanes thatare related to the traffic volume and design speed of the two lanehighway.
Table1: Width of traveled way of a two lane highway (AASHTO GreenBook, 2011) Preferable range of travel lane width according to the lane type isshown in Table 2. The width of travel lane for arterial roads is 3.6 m. However,lanes of 3.6 m cost more than 3.
3 or 3m for construction, they cost lessregarding maintenance. Due to wideness of the travel lanes on arterial roads, wheelconcentrations at the edge of pavement are reduced; thus maintenance cost for shoulders and pavement surface is reduced. In places where unequal lanewidths are used, locating the wider laneon the outside lane provides more space for large vehicles that usually occupythat lane, provides more space forbicycles, and allows drivers to keep their vehicles at a greater distance fromthe right edge. Table2: Recommended width of travel lane (AASHTO Green Book, 2011) Width of travel lane (m) Recommendations 2.7 – 3.6 Width of travel lane 3.6 High speed & high volume highway 3 Low speed facilities 2.
7 Used occasionally in urban areas if: • Traffic volume is low, and • There are extreme right-of-way constrains 3 Auxiliary lanes (like left, right, and U turn lanes) 3 – 4.8 Two Way Left Turn Lanes (TWLTLs) 2. Aimof the Study The aim of this study is to know the effects of lane width onspeed. The study focuses on vehicle speeds at different types of highways, andland use. The types of lanes of the highways in urban and rural areas are alsotaken into consideration in the study.3.
LiteratureReview Many studies have been done about the effects of lane width onspeed of vehicles. Rosey, et al. in 2009 studied the impacts of narrower widthof lanes on speed. Fixed-base simulator and real data were compared for knowingthe impacts of lane reduction on speeds. Thestudy was conducted in two steps: first, 43 drivers drove the simulators on twoconfigurations of lane widths (3 and 3.5 m); second, the results of thesimulator were compared with a field study, which was conducted by French fieldstudy carried out on a rural road. The results of the study showed that reducingthe lane width did not have impacts on vehicle speeds, but did induce theparticipants to drive closer to the center of the road.
It also showed thatoncoming vehicles induced subjects to move toward the right side of their lanes.Ibrahim, et al. in 2017 evaluated the effects of lane width androadside alignments on speed, lateral position, and possibility of comfortableovertaking in exclusive motorcycle lane.
Field experiment was conducted along apredetermined stretch of an exclusive motorcycle lane along the FederalHighway, Route 2 in the state of Selangor, Malaysia. The stretch over which theexperiment was conducted had a total length 20 km comprising both northboundand southbound with an average riding time of 20 min. The length of the roadway was selected inaccordance to the roadway characteristics, availability of a long tangentsection to detect overtaking events, and availability of suitable drop-off andpickup points. Along the selected route, average lane width was 3.05 m.Off-the-shelf DR- 9100 unit manufactured by Horiba Ltd was used as a data loggerto record real-time riding speed, video output, electrical signals and outputsfrom acceleration sensor simultaneously. Two cameras of high resolution werefixed on the center front and at the back of the motorcycle to providereal-time video images of the forward view and the area directly behind themotorcycle.
In an overtaking event, participants’ lateral position andproximity to the front motorcycle was estimated using ultrasonic distancesensor. Participants were 29 motorcyclists with average age of 25.6 years old.Descriptive statistics of overtaking events on tangents and curved sections oflane is shown in Table 3. Table 3: Descriptive statistics of all overtaking events (Ibrahim,et al., 2017) Overtaking events N=413 overtaking speed (km/h) Lane with (m) Maximum lateral acceleration (g) Average Std.Deviation Average Std.Deviation Average Std.
Deviation Tangential sections N =367 75.97 8.35 3.05 0.16 0.41 0.15 Curved sections N =46 64.
26 13.34 3.03 0.22 0.32 0.09 The overtaking speed on tangential sections of the roadway ishigher than the overtaking speed on curved sections.
A motorcyclist on curvedsections needs a greater distance and acceleration to reach overtaking speed onstraight sections. The researchers used several variables in the study, and thedetails of those variables are shown in Table 4. Table4: List of variables (Ibrahim, et al., 2017) Variable name Definition Categories Overtaking category Description of an overtaking event characterized by risk of collision 0. Not comfortable 1.Comfortable Lane width category Category of the lane width 2. Less than 3 m 3.
3 m or wider Roadside configuration Presence of guardrail on the roadside 0. No guardrail 1. Guardrail on the right side 2.
Guardrail on the left side Overtaking speeds Maximum speed achieved during overtaking continuous Front motorcycle’s (FM) lateral position, X1 Distance of front motorcyclist (FM) from the most left of edge line immediately before Continuous Participants’ proximity to the front motorcycle, X2 Distance of participants from the front motorcyclist (FM) during overtaking event continuous Participants’ lateral position, X3 Distance of participants from the most right of edge line during overtaking event continuous The results of the study showed that roadside configuration andlane width category tremendously affected participants’ overtaking speeds. Chitturi and Benekohal in 2005 investigated the effects of lanewidth on speeds of cars and heavy vehicles at work zones. Data werecollected at 11 work zones on Interstate highways in Illinois. One of the twotravel lanes in all of the work zones was closed. Four of the sites did nothave work activity. The width of the travel lane and the average free flowspeed (FFS) for each of the four sites were presented in Table 5. The informationregarding the four sites were the hourlyvolume, the percentage of trucks, the data used, the number of free-flowingvehicles, and the variance in the FFS at each of the sites. Informationpresented in Table 5 was used to obtain the values for speed reduction causedby the narrow lane widths in the work zones.
Table 5:Characteristics of different Sites (Chitturi and Benekohal, 2005) As indicated in the table,when the lane width was 16 ft in site 1, the Average FFS was 57.9 mph; however,when the lane width was 10.5 m in site 4, the FFS was 47.2 mph.
Regarding the effects of lane width on passenger cars, trucks, andall types of vehicles on FFS, Figure 1 shows the results.Figure 1: FFS with variation in lane width for different vehicletypes (Chitturi and Benekohal, 2005 It is obvious from the above figure that there is a constantdecrease in free flow speed as the lane width decreases from 16 to 10.5 ft. Whenthe lane width is 10.5 ft, the average value of the FFS of heavy vehicles wasnearly 45 mph, and for passenger cars was 48 mph. As the lane width increaseedfrom 11 to 12 ft, all vehicles speed increased with a constant rate. Figure 2 shows speed reductionin FFS for passenger cars, heavy vehicles, and all types of vehicles when thelane widths were 10.5 and 11 ft.
Figure 2: Observed reductions in FFSs of vehicles in work zones (Chitturiand Benekohal, 2005) Passenger cars FFS reduction was smaller than heavy vehicles for10.5 ft lane width. As the lane width increased to 11 ft, FFS for passengercars further decreased. Heavy vehicles FFS was reduced slightly compared to itsFFS reduction in 10.
5 ft. From the collected data of the site, it can beconcluded that there is a greater reduction in speed of all vehicles as thewidth of the lanes get narrower. The FFS reduction values recommended in HCM for basic freewaysections (4) and the reductions computed in this study were compared asshown in Table 7. Table 7: Observed Reductions in FFS for Work ZoneThe FFS at site 1 (lane width of 16 ft.) was 3.5 mph greater thanthe FFS at site 2. Therefore, the increase in FFS due to the wider travel lane wasset at 3.5 mph.
Similarly, the decrease in FFSs observed at sites 3 and 4 werethe speed reductions caused by the reduction of the lane width.The purpose of this comparison was to show that the HCM values couldnot be assumed to be applicable in work zones. HCM does not indicate anyincrease in FFS when the lane width is greater than 12 ft; therefore, a valueof 0 was used for the reduction recommended in HCM for a 16-ft lane. When thelane width was 11 ft, the observed reduction of 4.4 mph was 133 % more than thevalue of 1.
9 mph recommended in HCM. When the lane width was 10.5 ft, theobserved reduction of 7.2 mph was 69% greater than the value of 4.
25 mphrecommended in HCM. The conclusion is that the motorists reduced their speedsmore significantly in work zones than in a basic freeway section. When thesignificant difference between the work zone speed reductions and the HCM speedreductions were considered, it could be concluded that it would be wrong toassume that the values for a basic freeway section in HCM are applicable inwork zones.
Finally, it is recommended that the proposed values of FFSreduction can be used at work zones.Sharma, et al. in 2015 examined the “Safety and Operational Analysisof Lane Widths in Mid-Block Segments and Intersection Approaches in the Urban Environmentin Nebraska”. To determine lane width effects on operational analysis, linearregression models were used to examine lane width effects on the vehicles’speed at mid-block segments.
The Kolmogorov-Smirnov test was used as a statisticaltool to determine the effects of lane width on the vehicles’ headways in thequeue on the intersection approaches. The results of the study showed that laneswith 9 and 10 ft. wide had higher speed in the central business district at 25mph and areas outside of the central business district at 35 mph.
Lower trafficspeed was also observed in the areas with 40 mph and 45 mph speed limits forlane widths ranging from 9 to 10 ft. Dixon, et al. in 2015 conducted a study about “reducing lane andshoulder width to permit an additional lane on a freeway”. The research teamrecognized the operational and safety implications of using reduced lane andshoulder widths for a variety of freeway configurations.
Data including speed,crash, and geometric were used for freeways in Dallas, Houston, and SanAntonio. The research team observed an increase of about 2.2 mph in speed for a12-ft lane when compared to an 11-ft lane. Mecheri, et al. in 2017 studied “The effects of lane width,shoulder width, and road cross-sectional reallocation on drivers’ behavioraladaptations”.
The research team aimed at investigating lane and shoulder widtheffects on lane positioning strategies by examining vehicle distance from thecenter of the lane. The researchers also aimed to assess the impact of the lanewidth reduction, and shoulder widening on safety. The study was carried out ona fixed-base driving simulator to find out how speed and lateral position areaffected by different lane width and shoulder-width combinations when theroad’s cross-sectional width was reallocated. Participants of the experiment were 34 drivers. Mecheri, et al.concluded that reducing the width of the lane made participants drive closer tothe center of the road. Godley, et al. in 2004 studied the possibility driving speed reductionby using lane delineation.
A high-fidelity driving simulator was used on aseven two-lane rural roads with lane widths of 3.6, 3.0, or 2.5?m, and witheither a standard centerline (control), a wide painted hatched road centermarking, or a wide white gravel road center marking. Taking 28 experienceddrivers resulted in speed reduction on the narrowest lane width road. Godley,et al also concluded that the narrow lane width increased steering workload andreduced speeds through a speed-steering workload trade-off.
A new effectivemethod for reducing driving speed was proposed by the research team. The methodwas narrowing the lane width below 3.0?m by using a painted hatched road centermarking.
In 1990, Harwood investigated the” Effective Utilization of StreetWidth on Urban Arterials”. The researcher noticed that lane widths of 10feet or more resulted in accident rates. The results of the study showed thatnarrower lane widths (less than 11 ft.
) can be used successfully in urbanarterial street improvement projects where the additional space can be used to relievetraffic congestion or address specific accident patterns. The researcherrecommended the usage of 10 feet lane on sections where the street can’t bewidened to improve traffic operations or lessen specific accident patterns. Liu, et al. in 2016 tested the effects of lane width, lane positionand edge shoulder width on driving behavior for a three-lane underground urbanexpressway. In the experiment, a driving simulator was used with 24 volunteertest subjects. Five lane widths (2.
85, 3.00, 3.25, 3.50, and 3.75 m) and threeshoulder widths (0.50, 0.
75, and 1.00 m) were evaluated. The results from thestudy showed that lane and shoulder width had substantial effects on drivingspeed. Average driving speed increased from 60 km/h in the narrowest lane to 88km/h in the widest lane. The results of the study showed that the narrowerlanes and shoulders widths, reduced speed and lateral lane deviation areobtained. Besides the effect of lane width was greater than shoulder width. Chandra and Kumar in 2003 investigated the “Effect of Lane Width onCapacity under Mixed Traffic Conditions in India”.
Data were collected at tensections of two-lane roads in different parts of India. Total widths of the paved surface of the roadapart from its shoulders ranged from 5.5 to 8.
8 m. Vehicles were divided intonine different categories and after corresponding into passenger car units (PCU),were valued at each road sections. The capacity of a two-lane road increasedwith total width of the paved surface. Chandra and Kumar used the relationship forderiving the adjustment factors for substandard lane widths. 4. Conclusion This paper briefly reviewed the studies of other researchersregarding the effect of lane width on speed. The most significant outcomes are shownbelow:· Accordingto Rosey’s study, speed of the vehicles were not influenced by the width of thelane.· FromIbrahim studies, it was concluded that speeds of drivers were greatly affectedby width of the lane.
· Chitturiconcluded that reductions in FFS occur as the lane decreases from 16 to 10.5ft. The researcher proved that HCM reductions in FFS are only applicable forfreeway but not for work zones.· Accordingto Sharma study, lanes with 9 and 10 ft. wide had higher speed in the centralbusiness district at 25 mph and areas outside of the central business districtat 35 mph. · InDixon study, there was an increase of about 2.
2 mph in speed for a 12 ft. lanewhen compared to an 11 ft. lane. · In Mecheristudy, reduction in lane width caused participants drive closer to the centerof the road. · Godleyproposed a new method for reducing speed by narrowing the lane width below3.
0?m using a painted hatched road center marking. · Tenfeet lane width was recommended by Harwood on sections where the street can’tbe widened for further improvement of traffic operations or minimizing specificaccident patterns. · Liuconcluded that lane width of highways had an enormous impact on driving speed.The paper showed an increase in driving speed from 60 km/h in the narrowestlane to 88 km/h in the widest lane. · FromChandra studies, it was concluded that the capacity of a two-lane roadincreased with total width of the paved surface.
References· AmericanAssociation of State Highway and Transportation Officials. A Policyon Geometric Design of Highways and Streets. AASHTO,Washington, D.C. 2011.· Rosey,F., Auberlet, J. M.
, Moisan, O., & Dupré, G. (2009). Impact of narrowerlane width: Comparison between fixed-base simulator and real data. TransportationResearch Record: Journal of the Transportation Research Board, (2138),112-119.
· Ibrahim,M. K. A., Hamid, H.
, Law, T. H., & Wong, S. V. (2018). Evaluating the effectof lane width and roadside configurations on speed, lateral position andlikelihood of comfortable overtaking in exclusive motorcycle lane. AccidentAnalysis & Prevention, 111, 63-70.
· Chitturi,M., & Benekohal, R. (2005). Effect of lane width on speeds of cars andheavy vehicles in work zones.
Transportation Research Record: Journalof the Transportation Research Board, (1920), 41-48.· Sharma,A., Li, W., Zhao, M., & Rilett, L. (2015). Safety and Operational Analysisof Lane Widths in Mid-Block Segments and Intersection Approaches in the UrbanEnvironment in Nebraska.
· Dixon,K., Fitzpatrick, K., Avelar, R., Perez, M.
, Ranft, S. E., Stevens, R., …& Voigt, A. P.
(2015). Reducing lane and shoulder width to permitan additional lane on a freeway: Technical report (No.FHWA/TX-15/0-6811-1). Texas A & M Transportation Institute.· Mecheri,S.
, Rosey, F., & Lobjois, R. (2017). The effects of lane width, shoulderwidth, and road cross-sectional reallocation on drivers’ behavioraladaptations.
Accident Analysis & Prevention, 104,65-73.· Godley,S. T., Triggs, T. J., & Fildes, B. N.
(2004). Perceptual lane width, wideperceptual road centre markings and driving speeds. Ergonomics, 47(3),237-256.
· Harwood,D. (1990). NCHRP Report 330: Effective Utilization of Street Width. TransportationResearch Board, National Research Council, Washington, DC.· Liu,S., Wang, J.
, & Fu, T. (2016). Effects of lane width, lane position andedge shoulder width on driving behavior in underground urban expressways: Adriving simulator study. International journal of environmentalresearch and public health, 13(10), 1010.· Chandra,S., & Kumar, U. (2003). Effect of lane width on capacity under mixedtraffic conditions in India.
Journal of transportation engineering, 129(2),155-160.