Paper No.498 ‘‘CAPACITY ESTIMATION PROCEDURE FOR TWO-LANE ROADS UNDER MIXED TRAFFIC CONDITIONS” + By DR. SATISH CHANDRA* CONTENTS Page 1. Introduction ... ... 139 2. Objectives of Research ... ... 140 3. Background Literature ... ... 140 4. Data Collection and Research Methodology ... ... 142 5. Analysis of Data ... ... 146 6. Method of Capacity Estimation ... ... 166 SYNOPSIS Data collected at more than 40 sections of two-lane roads in different parts of the country are analysed. The effect of influencing parameters like gradient, lane width, shoulder width, traffic composition, directional split, slow moving vehicles and pavement surface conditions, on capacity of two-lane roads under mixed traffic conditions is evaluated and adjustment factors for each of these conditions are proposed. Based on these adjustment factors, a systematic procedure to evaluate capacity of a two-lane road under mixed traffic conditions is presented in this paper. 1. INTRODUCTION Two-lane highways compose the predominant portion of most National and State Highway system in the country. Two-third length of National Highways and more than 50 per cent length of State Highways is still two-lane wide. Traffic operation on a two-lane two-way highway is unique. Lane changing and overtaking are possible only in the face of on-coming traffic in the opposing lane. The overtaking demand increases rapidly as traffic volume increases, while passing opportunities in the opposing 139 + Written comments on this Paper are invited and will be received upto 31st December, 2004 * Associate Professor, Department of Civil Engineering, Indian Institute of Technology, Roorkee-247 667
Transcript
Paper No.498
‘‘CAPACITY ESTIMATION PROCEDURE FORTWO-LANE ROADS UNDER MIXED TRAFFIC
CONDITIONS” +
By
DR. SATISH CHANDRA*
CONTENTS
Page
1. Introduction ... ... 1392. Objectives of Research ... ... 1403. Background Literature ... ... 1404. Data Collection and Research Methodology ... ... 1425. Analysis of Data ... ... 1466. Method of Capacity Estimation ... ... 166
SYNOPSIS
Data collected at more than 40 sections of two-lane roads in different partsof the country are analysed. The effect of influencing parameters like gradient, lanewidth, shoulder width, traffic composition, directional split, slow moving vehiclesand pavement surface conditions, on capacity of two-lane roads under mixed trafficconditions is evaluated and adjustment factors for each of these conditions areproposed. Based on these adjustment factors, a systematic procedure to evaluatecapacity of a two-lane road under mixed traffic conditions is presented in this paper.
1. INTRODUCTION
Two-lane highways compose the predominant portion of most Nationaland State Highway system in the country. Two-third length of NationalHighways and more than 50 per cent length of State Highways is stilltwo-lane wide. Traffic operation on a two-lane two-way highway is unique.Lane changing and overtaking are possible only in the face of on-comingtraffic in the opposing lane. The overtaking demand increases rapidly astraffic volume increases, while passing opportunities in the opposing
139
+ Written comments on this Paper are invited and will be received upto 31stDecember, 2004
* Associate Professor, Department of Civi l Engineering, Indian Institute ofTechnology, Roorkee-247 667
DR. CHANDRA ON140
lane decline as volume increases. Therefore, flow in one direction influencesflow in the other direction. The problem is more acute in case of mixedtraffic flow when speed differential among different categories of vehiclesis quite substantial. It increases the desired number of overtakingconsiderably with limited opportunities to overtake.
Prediction and knowledge of capacity is fundamental in design,planning, operation and layout of road network sections. Roadway factorsthat influence capacity of a two-lane road include lane width, gradient,lateral clearance, width and type of shoulder. Lane and shoulder widthcan have a significant impact on traffic flow. Narrow lanes cause vehiclesto travel closer to each other laterally by slowing down or by observinglarge longitudinal gaps for a given speed. This effectively reduces thecapacity. Important traffic conditions that affect capacity of a two-laneroad are composition of traffic stream, directional split and presence ofslow moving vehicles in the stream. Environmental conditions such aswet pavement or snow and ice conditions, rain, darkness, fog, parkingregulation affect the driver performance and hence capacity. There areindications that wet or icy pavement can reduce capacity by 5-15per cent25.
2. OBJECTIVES OF RESEARCH
The present study was undertaken with the following objectives:
(i) To analyze traffic flow data collected on two-lane roads varyingin roadway and traffic conditions.
(ii) To study the effect of influencing parameters like gradient,lane width, shoulder width, traffic composition, directionalsplit and pavement surface roughness on capacity of two-laneroads under mixed traffic conditions.
(iii) To estimate adjustment factors for each parameter analyzed inthe present study.
(iv) To evolve a systematic procedure to determine capacity of atwo-lane road under mixed traffic conditions.
3. BACKGROUND LITERATURE
Taragin and Eckhardt (1953)23 studied the effect of shoulder onspeed and lateral placement of motor vehicles and found that when two-lane pavements on main highways are 6 m in width or less, shouldersshould be constructed with at least 1.2 m of stabilized material, adjacentto pavement plus additional width of grass and gravel. Leong (1978)16
measured speeds and capacity at 31 sites on rural highways in NewSouth Wales. The sites had varying lane and shoulder width and all sites
CAPACITY ESTIMATION PROCEDURE FOR TWO-LANE ROADS UNDER MIXED
TRAFFIC CONDITIONS141
had gravel shoulders. The data were analyzed using multiple regressionand it was suggested that speed increased with increasing shoulderwidth. Prakash (1970)18 also observed that the highway capacity isconsiderably influenced by the type and width of shoulder. Farouki andNixon (1976)8 studied the effect of carriageway width on speed of carsin the special case of free-flow conditions in sub-urban roads at Belfast.It was found that the mean free speed of cars in suburban area increaseslinearly with the carriageway width over a certain range of width (5.2 to11.3 m). Turner et al. (1982)24 found that the conversion of a shoulder toan additional travel lane could be expected to increase average-speed ofa two-lane highway by about 5 per cent for volumes exceeding 150veh/h. Yagar and Aerde (1983)26 found that speed changes exponentiallywith change in lane width. Chandra and Kumar (1996)5 studied the effectof shoulder condition on speed of different types of vehicles and theirplacement on road during passing and overtaking maneuvers on singleand two-lane highways. William and Reilly (1992)25 provided a summaryof operational techniques that can be used to improve level of serviceand capacity on two-lane highways. Ramanayya (1988)19 observed thatthe capacity standards adopted in western countries do not take intoaccount the mixed traffic characteristics prevalent in India. Sarna et al.(1989)21 emphasized on the need of developing highway capacity normsfor Indian highways. Kadiyali et al. (1991)13 observed that vehicle speedson Indian roads have increased during the past ten years. Speed-flowrelationships have also undergone changes. Pursula and Enberg (1991)reported from Finland that the highest flow rate measured on two-lanetwo-way road was 2500 veh/h with a directional split of 50/50. Fi (1994)9
reported that traffic characteristics were similar to HCM and expectedtraffic volume on two-lane highway was near 1500 pcu/h/l. Bang et al.(1995)1 developed speed-flow relationship and simulation model for two-lane road in Indonesia and found that free flow speed for two-lane roadsunder ideal conditions is considerably lower in Indonesia than in developedcountries. Sahoo et al. (1996)20 found that increase in traffic volumedecreases the speed of vehicles. Parker (1996)17 observed that knowledgeof traffic composition plays an important role in determining capacity.Kumar and Rao (1998)15 observed that speed density data could bereasonably represented by a linear relationship. Hossain and Iqbal (1999)11
studied vehicular free speed characteristics on two-lane national highwayof Bangladesh. Karan et al. (1978)14 developed relationship between averagespeed and pavement conditions for two-lane highways. Schofield (1986)studied effect of light and weather conditions on the speed and capacityof two-lane roads. Brilon and Ponzlet (1997)2 studied influences ofenvironmental factors on the speed-flow relationships on Germanautobahns.
DR. CHANDRA ON142
4. DATA COLLECTION AND RESEARCH METHODOLOGY
The data for this study were collected at more than 40 sections oftwo-lane roads to determine the effect of gradient, lane width, directionalsplit, shoulder’s condition and pavement roughness on capacity of two-lane roads. The sections were so selected that the effect of each parametercould be studied individually. The details of these sections are given inTables-1 to 4.
TABLE-1. DETAILS OF SECTIONS SELECTED FOR EFFECT OF GRADIENT
CAPACITY ESTIMATION PROCEDURE FOR TWO-LANE ROADS UNDER MIXED
TRAFFIC CONDITIONS143
TABLE-3. SECTIONS SELECTED FOR DIRECTIONAL SPLIT AND PERCENT SMV
Section Section Width of Directional % SMVNo. carriageway Split
(m)
16. km–12 of NH-30 7.0 55:45 7.1
17. km–26 of NH-30 7.0 62:38 15.6
18. km–65 of NH-31 7.0 57:43 5.7
19. km–38 of NH-58 7.0 100:0 12.7
20. km–135 of NH-58 7.0 53:47 10.9
21. km–39 of NH-73 7.0 59:41 21.9
22. km–2 of NH-77 7.0 62:38 28.4
23. km–5 of NH-83 7.0 54:46 22.0
24. km–5 of SH-26 (Punjab) 7.0 51:49 12.6
TABLE -4. SECTIONS SELECTED FOR SHOULDER CONDITION
Section Name of the Carriageway Shoulder Physical condition
No. Road Width (m) Type Width of shoulder(m)
25. km–3 of NH-19 7.0 Brick on 1.6 Drop at pavement(Hazipur-Chhapra edge edge was 5.0 cm.Road) No depression on
shoulder was present.26. km–15 of NH-19 7.0 Earthen 1.6 Drop at pavement
edge was 9.0 cm.Potholes up to 5.0 cmwere present onshoulder.
27. km–8 of NH-19 7.0 Earthen 1.5 Drop at pavementedge was 12.5 cm.Depressions onshoulder were up to10-15 cm.
28. km–155 of NH-30 7.0 Earthen 1.2 Drop at pavement(Patna-Bihta-Ara edge was 5.0 cm.Road) Depressions on
shoulder were up to5.0 cm.
29. km–134 of NH-30 7.0 Earthen 1.6 Drop at pavementedge was 8.0 cm.Depressions inshoulder were up to3.0 cm.
DR. CHANDRA ON144
30. km–142 of NH-30 7.0 Earthen 1.8 Drop at pavementedge was 12.0 cm.Depressions onshoulder were up to5.0 cm.
31. km–17 of NH-31 7.0 Soil 1.8 Drop at pavement(Barhi-Bakhtiarpur- Gravel edge was 5.0 cm.Barauni-Khagariya mix No Depression inRoad) shoulder area.
32. km–24 of NH-31 7.0 Soil 1.8 Drop at pavementGravel edge was 9.0 cm.
mix Depressions onshoulder were up to5.0 cm.
33. km–28 of NH-31 7.0 Soil 1.3 Drop at pavement Gravel edge was 15.0 cm.
mix Depressions onshoulder were up to10.0 cm.
Section Name of the Carriageway Shoulder Physical condition
No. Road Width (m) Type Width of shoulder(m)
The data were collected by video filming technique except at sectionsgiven in Table-4 where manual method was adopted. A longitudinal trapof 25-30 m was made on the road and recording was done for 3-4 hoursduring morning/evening hours of a typical weekday. To study the effectof shoulder condition on placement of vehicles, the data on speed,placement and volume of vehicles was collected manually at each section.The pavement width was divided in sections of 10 cm using white self-adhesive cloth tape. These sections were marked seriatim in white color.Movement of all categories of vehicles was observed and position of leftwheels of a vehicle passing though the section was recorded. The smallsize vehicles like two wheelers or bicycles do not require much use ofshoulder and therefore these vehicles were not included in data collectionfor placement. In order not to influence the driver, the recording wasdone by an observer standing at an obscure place such as the shadowof a roadside tree at a little distance away from the pavement. For thesame reason of not disturbing the general pattern of traffic, observationswere restricted to a typical weekday and clear sky. The following vehicleswere considered for placement of vehicles from the edge of the pavement:
(i) Car and jeep(ii) Heavy vehicle (HV like bus and truck)
TABLE -4. SECTIONS SELECTED FOR SHOULDER CONDITION (CONTD.)
CAPACITY ESTIMATION PROCEDURE FOR TWO-LANE ROADS UNDER MIXED
TRAFFIC CONDITIONS145
(iii) Car-car passing or overtaking and placement was observed forcar left to the observer.
(iv) Car-HV passing or overtaking and placement was observedfor vehicle left to the observer.
(v) HV-HV passing or overtaking and placement was observed forHV (Heavy Vehicle) left to the observer.
Speed data of car, LCV (Light Commercial Vehicle) and HV duringfree moving and passing or overtaking maneuvers are given in Table-5.The placements of vehicles from pavement edge during free movement ofcar and heavy vehicles (HV) and during passing/overtaking of car-car,car-HV and HV-HV are given in Table-6.
TABLE-6. PLACEMENT OF VEHICLES FROM PAVEMENT EDGE
Section Weidth Placement (cm of single Placement (cm) of vehicle No. of Pave- vehicle from pave- during crossing/over-
No. Category Vehicles Average Projectedincluded Dimensions Rectangular
(m) Area (m2)
1. Car Car, Jeep, Van 3.72 x 1.44 5.362. Bus Bus 10.1 x 2.43 24.543. Truck Truck 7.5 x 2.35 17.484. LCV Mini bus, large vans, 6.1 x 2.1 12.81
tractors5. Thee Wheelers Thee Wheelers 3.2 x 1.4 4.486. Two Wheelers Scooter, Motorbike 1.87 x 0.64 1.20
Mopeds7. Cycle Bicycles 1.9 x 0.45 0.868. Rickshaw Pedal rickshaw, carts 2.7 x 0.95 2.569. Tractor Tractor trailer 7.4 x 2.2 16.2810. Animal driven Bullock-cart, Horse cart 5.5 x 1.75 9.63
vehicle (ADV)
5.1. Estimation of Equivalency Factors
The main problem in developing the analytical speed-flow relationship isheterogeneity of traffic. The vehicles in the mix produce different impedance dueto their varied static and dynamic characteristics. Hence simply adding thenumber of vehicles does not give the authentic speed flow relationship. For thisreason, the vehicles are normally presented in terms of standard type of vehicleusing certain conversion factors. Generally, passenger car is adopted as standardvehicle and therefore the factor is known as passenger car unit (PCU). Manyresearchers have developed methods to estimate PCU for a vehicle type. Theinteresting point to note is that each of these studies has resulted into differentPCU values for the same type of vehicle. There exists large variation in PCUvalues being adopted in different parts of the world.
In the present study, the PCUs are calculated as follows. The basicconcept used to estimate the PCU is that it is directly proportional to the ratioof clearing speed, and inversely proportional to the space occupancy ratio withrespect to the standard design vehicle, a car, i.e.
Speed ratio of the car to the ith vehiclePCUi = _______________________________________________________ (Eqn. 1)
Space ratio of the car to the ith vehicle
Where,PCUi = passenger car unit value of ith type vehicle
5. ANALYSIS OF DATA
The data were analyzed to study the effect of influencing param-eters on capacity of two-lane roads. All vehicles were divided into 10categories as shown in Table-7.
CAPACITY ESTIMATION PROCEDURE FOR TWO-LANE ROADS UNDER MIXED
TRAFFIC CONDITIONS147
Speed ratio of the car to the ith vehicle = Vc/V
i
Space ratio of the car to the ith vehicle = Ac/A
i
Vc
= speed of car (km/h)V
i= speed of ith type vehicle (km/h)
Ac
= static (projected rectangular) area of a car (m2)A
i= static (projected rectangular) area of ith type of vehicle (m2)
Therefore,
Vc/V
i
PCUi = __________ (Eqn. 2)A
c/A
i
The first variable of speed ratio in Equation (2) will be the functionof composition of traffic stream as the speed of any vehicle type dependsupon its own proportion and type and proportions of other vehicles.Hence speed of any vehicle type will be true representation of overallinteraction of a vehicle type due to presence of other vehicle of its owncategory and of other types. The second variable represents thepavement occupancy with respect to car.
The PCU values for different categories of vehicles were computedat various sections and these are given in subsequent sections.
5.2. Effect of Grade
The PCU values for different types of vehicles with grade at differentsections are given in Table-8. Further discussion on these values is givenelsewhere (Chandra and Goyal, 2001)4. The PCU values given in Table-8were used to convert all vehicles into equivalent number of passengercars and speed–volume relationships were plotted. The capacity values
TABLE -8. PCU FOR DIFFERENT TYPES OF VEHICLES AT DIFFERENT GRADIENT
The purpose of this part of study was to estimate capacity of two-lane roads with varying carriageway width. Data collected at sectionsgiven in Table-2 were analyzed to determine the effect of lane width.Table-10 presents the PCU values as derived at these sections usingEquation (2). Figs. 1 to 3 show the variation in PCU for different typesof vehicles with varying lane width at different sections.
TABLE-10. PCU FOR DIFFERENT TYPES OF VEHICLES ON VARYING CARRIAGEWAY -WIDTH
The typical speed-volume relationship drawn to estimate capacity isshown in Fig. 4. Similar curves were drawn at other sections also. Thecapacity values estimated from these curves are given in Table-11.
with grade at different sections are given in Table-9. It indicates that eachper cent of upgrade decreases the capacity by 2.61 per cent and eachper cent of downgrade increases the capacity by 3.09 per cent.
CAPACITY ESTIMATION PROCEDURE FOR TWO-LANE ROADS UNDER MIXED
TRAFFIC CONDITIONS149
Fig. 1. Effect of carriageway width on PCU for large vehicles
Fig. 2. Effect of carriageway width on PCU for cycle & 2-wheeler
DR. CHANDRA ON150
Fig. 3. Effect of carriageway width on PCU for 3-wheeler & rickshaw
Fig. 4. Speed-volume relation at Section-11
CAPACITY ESTIMATION PROCEDURE FOR TWO-LANE ROADS UNDER MIXED
TRAFFIC CONDITIONS151
TABLE-11. CAPACITY OF TWO-LANE ROAD WITH DIFFERENT CARRIAGEWAY WIDTH
Name of the Road Carriageway width (m) Capacity (pcu/h)
Where,C = Capacity of road (pcu/h)w = Total width of the carriageway (m)
DR. CHANDRA ON152
The values given in parentheses are the ‘t’ values of coefficients, which aresignificant at 5 per cent level. Thus the capacity of a 7.2 m wide carriageway is 2818pcu/h, which is quite close to the value suggested in HCM (1994) at equal directionalsplit. Adjustment factors for other lane widths are given in Table-12.
TABLE-12. COMPARISON BETWEEN LANE WIDTH ADJUSTMENT FACTORS
Serial No. Lane-width (m) HCM, 1994 Swedish, 1977 Present study
5.4. Effect of Directional Split and Slow Moving Vehicles
The nine sections given in Table-3 were analysed to study theeffect of directional split and slow moving vehicles (SMA) on capacityof two-lane roads. All the sections had 7.0 m wide carriageway. Thedetails of directional split and percent slow moving vehicles at each ofthese sections are also given in Table-3. The PCU values as estimatedusing equation (2) at different sections are given in Table-13. The widevariation in PCU factors at different sections is attributed to the changein traffic conditions. These values were used to develop speed-volumerelationships. The typical curve is shown in Fig. 6. The capacity valuesas estimated at different sections are given in Table-14.
TABLE-13. PCU FACTORS AS ESTIMATED AT DIFFERENT SECTIONS
The results given in Table-14 show an increase in capacity with thedecrease in percentage of slow moving vehicles in traffic stream. Further,the capacity decreases as the directional split moves away from an evensplit of 50:50. For understanding the effect of slow moving vehicles(SMV) and directional split separately, it is essential to keep one of them
TABLE-13. PCU FACTORS AS ESTIMATED AT DIFFERENT SECTIONS (CONTD.)
DR. CHANDRA ON154
TABLE-14 . CAPACITY VALUES AT DIFFERENT DIRECTIONAL SPLIT AND PERCENT SMV
Section Identification Directional Percent Show CapacitySplit Moving Vehicles (pcu/h)
km 12 of NH-30 55:45 7.1 2858km 26 of NH-30 62:38 15.6 2647km 65 of NH-31 57:43 5.7 2322km 38 of NH-58 100:0 12.7 2283km 135 of NH-58 53:47 10.9 2803km 39 of NH-73 59:41 21.9 2132km 2 of NH-77 62:38 28.4 2144km 5 of NH-83 54:46 22.0 2270km 5 of SH-26 51:49 12.6 2743
constant and study the effect of other one. For practical reasons, thepercentage slow moving vehicle (% SMV) from 10.9 per cent to 15.6 percent is taken constant while studying the effect of directional split.Similarly, directional split of 51:49 to 55:45 is considered constant forstudying the effect of SMV on capacity of two-lane roads. The resultsare plotted in Fig. 7. The effect of directional split is linear. It is givenby equation (4)
C = 3398 – 9.53 QP
(Eqn. 4) (79.76) (8.91)
where, C is the capacity in pcu/h and QP is the per cent traffic in main
direction.
Fig. 7. Effect of directional split on capacity
CAPACITY ESTIMATION PROCEDURE FOR TWO-LANE ROADS UNDER MIXED
TRAFFIC CONDITIONS155
The effect of slow moving vehicles on capacity of two-lane roadsis shown in Fig. 8. The trend is expressed by the following equation.
TABLE -15. ADJUSTMENT FACTORS FOR DIRECTIONAL SPLIT
Spl i t Capacity from Equation (4) Adjustment factor
The values given in parentheses are the ‘t’ values of coefficients,which are significant at 5 per cent level. It gives he capacity of a two-lane road at 50:50 split as 2790 pcu/h, which is similar to the value givenin HCM (1994). The adjustment factors for different splits are given inTable-15.
Fig. 8. Effect of SMV on capacity
C = 2920 + 24.5 S – 2.2 S2 (Eqn. 5) (327.35) (14.16) (13.36)
Where S is the percentage slow moving vehicles in the trafficstream. The values given in parentheses are the ‘t’ values of coefficients,which are significant at 5 per cent level. The adjustment factors for SMVare given in Table-16.
DR. CHANDRA ON156
TABLE -16. ADJUSTMENT FACTORS FOR SMV
SMV (%) Capacity as per equation Adjustment factor(6), pcu/h
It is assumed that vehicles will make use of full width of pavementif good types of shoulders are provided on either side of the road.Although, the data were collected at locations where shoulders weredamaged, some data were collected at locations having good conditionof shoulders also. The 9 sections as given in Table-4 were selected fordata collection. The pavement riding quality was almost same at all thesections but they were varying in shoulder conditions.
5.5.1. Classification of shoulders
In order to have better appreciation of results, the shoulders weredivided into different categories depending upon their physical conditionat time of data collection. The method suggested by Chandra and DevRaj (1999)6 was adopted for objective assessment of shoulders. Thecriteria of this classification is given below:
(i) Good: The shoulders on either side are properly maintained andminor or no settlement of gravel or earthen shoulders (less than25 mm in depth) with the pavement edge. Shoulders can be usedat reasonably high speed.
(ii) Average: Drop at pavement edge is 25-50 mm. Broken portion ofthe shoulders is less than 25 per cent. Shallow potholes are presentbut the shoulder can be used at low speed.
(iii) Poor: Drop at pavement edge is 50-100 mm. 25-50 per cent shouldersare either broken or removed. Deep potholes are formed in thesurface. The shoulder can be used at considerably low speed only.
(iv) Bad: Drop at pavement edge is greater than 100 mm and more than50 per cent shoulders are broken. Deep potholes in the shouldersurface are present. It cannot be used by vehicle even at low speeddue to danger of overturning of vehicles.
CAPACITY ESTIMATION PROCEDURE FOR TWO-LANE ROADS UNDER MIXED
TRAFFIC CONDITIONS157
Based on the above classification, the categories of various sectionsselected for study are given in Table 17.
TABLE -17. CATEGORIES OF SHOULDER AT VARIOUS SECTIONS
Category of Shoulder Sections
Average 25, 28, 31Poor 26, 29, 32Bad 27, 30, 33
5.5.2. Loss in width of carriageway
Physical condition of shoulders plays an important role for thedevelopment of full traffic capacity of a highway. The lateral placementof vehicles and thus utilisation of full carriageway width depends uponthe width of carriageway and type and condition of shoulder. The fielddata given in Tables-5 and 6 indicate that the vehicles preferred to lowerdown their speed rather than coming on to the shoulders, which arepoorly maintained. The loss in width of carriageway for different conditionsis given in Table-18. As may be seen, the per cent loss in the width ofthe carriageway varies from 36.0 to 52.3 per cent for single moving vehicledepending upon the type of vehicle involved. During passing/overtaking,the percent loss in width of carriageway varies 18.0 to 35.4 per centdepending upon the type of vehicle involved in passing/overtakingmanoeuvres and condition of shoulder. At all sections, the loss in widthof carriageway for movement of a single car is greater than that for themovement of a HV and the loss in width of the carriageway during car-car passing/overtaking manoeuvres is more than that during HV-HV passing/overtaking manoeuvres. Table-18 indicates that the loss in the carriagewaywidth increases progressively as the condition of shoulder deteriorates,which directly affect the capacity of the road. Figs. 9 & 10 show theplacement of a single vehicle for different conditions of shoulders. Withaverage condition of shoulder about 50 per cent of cars and 50 per centof HVs had their placement within 115 cm and 65 cm respectively, frompavement edge while for bad condition of shoulder, only 29 per cent carsand 10 per cent HVs had their placement within these limits.
DR. CHANDRA ON158
TABLE-18. LOSS IN WIDTH OF CARRIAGEWAY RELATED TO CONDITION OF SHOULDER
Section Per cent loss in Per cent loss in effective with of Shoulder No. effective width of carriageway during passing/ Condit ion
CAPACITY ESTIMATION PROCEDURE FOR TWO-LANE ROADS UNDER MIXED
TRAFFIC CONDITIONS159
Fig. 10. Placement of heavy vehicle on two-lane road
5.5.3.Reduction in speed during passing/overtaking manoeuvres
Spot speeds of vehicles were measured during passing/overtakingmanoeuvres and during single free moving vehicle to estimate the averagespeed of each type of vehicle on a section. The percent reductions inspeed of individual vehicle during above condition are given in Table-19.Two wheeler and bicycles were found to be unaffected by crossingvehicles due to their small size but other vehicles were forced to reducetheir speed due to poor maintenance of shoulders. The maximum reductionin speed of a car while crossing another car is 10.7 per cent. Similarly, forthe same condition of shoulder, maximum reduction in speed of a HVwhile crossing another HV is 28.4 per cent. It indicates that the vehiclesreduce their speed considerably while crossing another vehicle. Forgood condition of shoulder the reduction in speed is very marginal,whereas for bad condition of shoulder it is quite substantial. The reductionis due to shoulder’s influence on the drivers’ behaviour during crossing.It should be noted here that the combined effect of speed reduction ofindividual vehicle and the loss in carriageway width on capacity will bemuch more significant.
DR. CHANDRA ON160
TABLE -19. REDUCTION IN SPEED DURING PASSING/OVERTAKING MANOEUVRES
Sec- Pave- Per cent reduction Per cent reduc- Per cent reduc- Shouldert ion ment in speed of car tion in speed tion in speed condi-No. width during crossing a of LCV during of HV during t ion
Data collected manually at 3 sections were used to develop speed-volume relationships. The traffic volume was converted into equivalentnumber of passenger cars and the mean speed was calculated. The freestream speed and capacity values as derived from these curves are givenin Table-20.
TABLE-20. MEAN FREE SPEED AND CAPACITY FOR DIFFERENT CONDITION OF SHOULDERS
Section Conditon of Mean Free CapacityNo. Shoulder Speed (km.h) (pcu/h)
31. Average 64.2 265432. Poor 60.7 242733. Bad 55.7 1831
The mean free speed of traffic stream changes from 64.2 km/h to55.7 km/h as the shoulder condition changes from average to bad. Thecapacity values as calculated for 3 sections are also given in Table-20.The capacity of a two-lane road with average condition of shoulder(section 31) is 2654 PCU/h. It reduces to 1831 PCU/h (section 33) whenshoulder was in bad condition.
CAPACITY ESTIMATION PROCEDURE FOR TWO-LANE ROADS UNDER MIXED
TRAFFIC CONDITIONS161
5.6. Effect of Road Roughness
This part of study was aimed at determining the effect of roadroughness on free flow speed (FFS) and capacity of two-lane roads.Therefore, the data were collected in two parts. In the first part, theobservations were taken for road roughness and free flow speed ofvehicles. The roughness measurements were taken on sections, free fromthe effect of curvature, gradient and intersections. The shoulders were ingood conditions and road surface was not cracked or pot holed. Theroughness was measured using British Towed fifth wheel bump integratoron the following sections of the two-lane highways.
(a) National Highway (NH) – 73 in the state of U.P.(b) State Highway (SH) – 59 in the state of Uttaranchal(c) National Highway (NH) - 7 in the state of Andha Pradesh
18 km length of NH-73, 5 km of SH-59 and 33 km of NH-7 wasselected for roughness and speed measurements.
Free flow speed of a vehicle was measured using radar gun at every100 m interval of a kilometre and the average speed for the entire kilometrewas determined. Only 2 categories of vehicles viz. car and heavy vehicles(bus & truck) were considered for this part of study. About 10 vehiclesof each category were observed for speed on each location.
In the second part of the study, speed-volume data were collectedat 8 locations of two-lane highways. All sites were straight, level and freefrom any restriction to traffic movement. The video recording techniquewas used to collect the data for 4-5 hours on a typical weekday and thesewere analysed to plot speed-volume diagram.
5.6.1. Effect of roughness on free speed
The variation in flow free speeds of two principal vehicle categoriesviz. car and heavy vehicle with roughness is shown in Fig. 11. The freeflow speed (FFS) decreases with roughness according to the followingrelationships.
Fig. 11. Free flow speed related to road roughness
Where speed is in km/h and UI is unevenness index in mm/km. Thevalues given in parentheses are the ‘t’ values of coefficients, which aresignificant at 5 per cent level.
5.6.2. Effect of roughness on passenger car unit (PCU)
The PCU of a vehicle type was calculated by Equation (2). Theseare given in Table-21. Figs. 12 to 14 show the variation in PCU fordifferent types of vehicles with road roughness at different sections. Asmay be seen, the PCU for a vehicle type decreases linearly with roughness,the slope of linearity depends on the type of vehicle. Non-motorizedvehicles are not much influenced by road roughness while speed ofmotorized vehicles is greatly influenced. The fast moving vehicles aresubstantially influenced by road roughness while its effect on all othervehicle types is relatively low.
5.6.3. Effect of roughness on capacity
The PCU values given in Table-21 were used to convert heterogeneoustraffic stream into a homogeneous equivalent. The mean stream speedwas plotted against traffic volume. The capacity values as estimated fordifferent sections are given in Table-22.
CAPACITY ESTIMATION PROCEDURE FOR TWO-LANE ROADS UNDER MIXED
TRAFFIC CONDITIONS163
TABLE -21. PCU FOR DIFFERENT TYPES OF VEHICLE ON VARIOUS SECTIONS
Sec- UI PCUtion (mm/ Bus Truck LCV MAT Tractor 2- 3- Cycle ADV
Where,C = capacity in pcu/h.UI = unevenness index in mm/km
DR. CHANDRA ON166
The values given in parentheses are the ‘t’ values of coefficients,which are significant at 1 per cent level. The above relationship indicatesthat capacity of a two-lane road decreases by 300 pcu/h when roadroughness increases by 1000 mm/km.
6. METHOD OF CAPACITY ESTIMATION
HCM (2000)10 state that the capacity of a two-lane rural highway is1700 pcu/h for each direction of travel and normally does not exceed 3200pcu/h for both direction of travel combined. IRC:64-199012 indicates thedesign service volume of 15000 pcu/day for a two-lane road in plainterrain, which corresponds to a capacity value of 2400-3000 pcu/h.The capacity of a two-lane road as estimated in the present study is 3140pcu/h under the following conditions.
(i) Lane width is 3.6 m.
(ii) Shoulders are in good condition and 1.2 m on either side.
(iii) The road surface is in good condition with surfaceunevenness of 2500 mm/km.
(iv) The directional split is 50:50.
(v) The section is levelled and straight.
(vi) Slow moving vehicles in the traffic steam are less than10 per cent.
It is, therefore, believed that the capacity of 3200 pcu/h as suggestedin HCM (2000)10 is achievable in India also. However, it is suggestedthat the capacity of a two-lane road in mixed traffic condition be takenas 3100 pcu/h. This value is affected by gradient, lane width, shoulderwidth, shoulder condition, slow moving vehicles, directional split, androad roughness. The capacity of a two-lane road under mixed trafficconditions is given by Equation (A).
Ca = c
b.f
g.f
w.f
ds.f
smv.f
s.f
ui(Eqn. A)
where,C
a= actual capacity under prevailing roadway and traffic conditions
Cb= basic capacity (3100 pcu/h)
fg, f
w, f
ds, f
smv, f
s, and f
UI are the adjustment factors for gradient, lane width,
directional split, slow moving vehicles, shoulder conditions and unevenness
CAPACITY ESTIMATION PROCEDURE FOR TWO-LANE ROADS UNDER MIXED
TRAFFIC CONDITIONS167
index. These adjustment factors are given below:
(i) Adjustment for grade: Every one percent upgrade decreases thecapacity by 2.60 per cent (f
g = 0.974) and one percent downgrade
increases the capacity by 3.0 per cent ((fg = 1.03)
(ii) Adjustment for lane width: Substandard lane width will reduce thecapacity as given in Table-A.
TABLE -A. RECOMMENDED CAPACITY REDUCTION FACTORS FOR SUB-STANDARD LANE WIDTH
Lane width (m) 3 .6 3 .3 3 .0 2 .7
Factor (fw) 1 .0 0.88 0.76 0.63
(iii) Directional split: If the flow is not balanced in two directions, thecapacity of the road reduces as per Table-B.
TABLE -B. RECOMMENDED FACTORS FOR DIRECTIONAL SPLIT
Flow in main direction (%) 50 60 70 80 90 1 0 0
Adjustment factor (fds
) 1 .0 0.97 0.94 0.90 0.87 0.84
(iv) Slow Moving Vehicles: The effect of slow moving vehicles isnegligible till their proportion in traffic stream is less than or equalto 10 per cent. After that their effect is to reduce the capacity asper Table-C.
TABLE -C. ADJUSTMENT FACTORS FOR SLOW MOVING VEHICLES (SMV)
SMV (%) £ 10 15 20 25 30
Adjustment factor (fsmv
) 1 .0 0.96 0.87 0.74 0.57
(v) Shoulder condition: The capacity of a two-lane road can be increasedby providing shoulders in good condition so that these can be usedby moving vehicles in emergency. Unfortunately, maintenance ofshoulder in India is extremely poor and in some case they becomereasons of accidents. For the purpose of analysis, shoulders aredivided in four categories as discussed in Section 5.5.1. Adjustmentfactor as derived here for these conditions of shoulders are givenin Table-D.
DR. CHANDRA ON168
TABLE -D- ADJUSTMENT FACTORS FOR SHOULDER CONDITION
Shoulder condi t ion Adjustment factor (f s)
Good 1.0Average 0.85
Poor 0.77Bed 0.58
(vi) Surface Unevenness: The surface unevenness has direct effect onoperat ing speed and thus on capacity. IRC:SP-30-1993 hasrecommended roughness values for different types of surfacing. Agood road surface is expected to provide full development of trafficcapacity. The capacity of a two-lane road reduces by 300 pcu/hwhen its road roughness reduces by 1000 mm/km.
ACKNOWLEDGEMENT
The present research was carried out under an AICTE sponsoredresearch scheme Capacity Analysis of Two-lane Roads under MixedTraffic Conditions. The financial assistance received from AICTE, NewDelhi for collection of traffic data for this project is gratefully acknowledged.
REFERENCES
1. Bang, K.L., Carlsson, A. and Palgunadi (1995), “Development of Speed FlowRelationship for Indonesia Rural Roads Using Empirical Data and Simulation”,Transportat ion Research Record 1484, Transportat ion Research Board,Washington D.C., pp. 24-32.
2. Brilon, W. and Ponzlet, M. (1997), “Variability of Speed-Flow Relationshipon German Autobahns” , TRR 1555, Transpor ta t ion Research Board,Washington, D.C., pp. 91-98.
3. Chandra, S. (1997), “Capacity of Urban Roads - Towards Review of IRC:106-1990”, Highway Research Bulletin, Highway Research Board, Indian RoadsCongress, New Delhi, pp. 43-70.
4. Chandra, S. and Goyal, N. K., (2001), “Effect of Grade on Capacity of Two-Lane Road”, HB No. 64, IRC, New Delhi, pp. 77.
5. Chandra, S. and Kumar, P. (1996), “Effect of Shoulder Condition on HighwayCapacity”, Proceedings International Seminar on Civil Engineering Practicesin Twenty First Century, Roorkee, India, pp. 512-519.
6. Chandra, S. and Raj, Dev (1999), “Roles of Shoulders in Traffic Operation”,Indian Highways, Vol. 27, No. 11, pp. 31-38.
CAPACITY ESTIMATION PROCEDURE FOR TWO-LANE ROADS UNDER MIXED
TRAFFIC CONDITIONS169
7. Chandra, S. and Sikdar, P.K. (2000), “Factors Affecting PCU in MixedTraffic Situations in Urban Roads”, Road Transport Research, Vol. 9, No. 3,Australian Road Research Board, pp. 40-50.
8. Farouki, O.T. and Nixon, W.J. (1976), “The Effect of Width of Sub-UrbanRoads on the Mean Free Speeds of Cars”, Traffic Engineering + Control, Vol.17 (12), London, pp. 518-519.
9. Fi, I. (1994), “Highway Capacity and Level of Service in Highway: Experiencewith Intersection without Traff ic Signals”, Country Reports of SecondInternational Symposium on Highway Capacity, Australian Road ResearchBoard Ltd.
10 . “Highway Capacity Manual” (2000), Transportation Research Board, SpecialReport 209, fourth edition, Washington D.C.
11 . Hossain, M. and Iqbal, G.A. (1999), “Vehicular Headway Distribution and FreeSpeed Characteristics of Two-lane Two-way Highway of Bangladesh”, Institutionof Engineers (India) Vol. 80, pp. 77-80.
12 . IRC:64 (1990), “Guidelines on Capacity of Roads in Rural Area”, (FirstRevision), Indian Code of Practice, Indian Roads Congress, New Delhi.
13 . Kadiyali, Dr. L.R., Lal, Dr. N.B., Sathyanarayana, M. and Swaminathana,A.K. (1991), “Speed-Flow Characteristics on Indian Highways”, Indian RoadsCongress Journal, Vol. 52-2, New Delhi, pp. 233-262.
14 . Karan, M.A., Ralph, H. and Kher, R. (1978), “Effect of Pavement Roughnesson Vehicle Speed”, TRR 602, Transportation Research Board, National ResearchCouncil, Washington, D.C., pp. 122-127.
15 . Kumar, V.M. and Rao, S.K. (1998), “Studies on Speed-Density-Flow Relationshipon a Few Stretches of NH-5 and NH-6”, Indian Highways, Vol. 26 (12), IndianRoads Congress, New Delhi, pp. 33-41.
16 . Leong, H.J.W. (1978), “Distribution and Trend of Free Speeds on Two-LaneTwo-Way Rural Highways in New South Wales”, ARRB 4, Part 1, pp. 798-814.
17 . Parker, M.T. (1996), “The Effect of Heavy Goods Vehicles and FollowingBehavior on Capacity at Motorway Sites.” Traffic Engg + Control, Vol.37(9), London, pp. 524-532.
18 . Prakash, V. (1970), “Highway Shoulder”, Journal of Indian Roads Congress,Vol. 33-3, pp. 441-446.
19 . Ramanayya, T.V. (1988), “Highway Capacity Under Mixed Traffic Conditions”,Traffic Engineering and Control, Vol. 29, No. 5, United Kingdom, pp. 284-300.
20 . Sahoo, P.K., Rao, S.K. and Kumar, V.M. (1996), “A Study of Traffic FlowCharacteristics on Two Stretches of National Highway No. 5” Indian Highways,Vol. 24 (4), Indian Roads Congress, New Delhi, pp. 11-17.
DR. CHANDRA ON170
21. Sarna, A.C., Jain, P.K. and Chandra, G. (1989), “Capacity of Urban Roads- A Case Study of Delhi and Bombay”, Highway Research Bulletin, No. 4,Indian Roads Congress, New Delhi, pp. 1-38.
22 . Shivananda, M. (1994), “Private Sector Participation in Development ofHighway Network of India”, Indian Highways, Indian Roads Congress, NewDelhi.
23 . Taragin, A and Eckhardt. H.G. (1953), “Effect of Shoulders on Speed andLateral Placement of Motor Vehicles”, HB Proceedings, Vol. 32, pp. 371-82.
24 . Turner, D.S., Rogness, R.O. and Fambro, D.B. (1982), “Shoulder UpgradingAlternatives to Improve the Operational Characteristics of Two-lane Highway”,TRB Annual Meeting, Washington, D.C.
25 . William, R. and Reilly, P.E. (1992), “Operational Aspect of Highway Capacity”,Traffic Engineering Hanbook, 4th Edition, Institute of Transportation Engineers,Prentice Hall, Englewood Cliffs, New Jersey, pp. 117-153.
26 . Yagar, S. and Aerde, M.V. (1983), “Geometric and Environmental Effects onSpeeds of 2-Lane Highways”, Transportation Research-A, Vol. 17A, No. 4,pp. 315-325.
CAPACITY ESTIMATION PROCEDURE FOR TWO-LANE ROADS UNDER MIXED
