t I t I

t \u.ll

slJb'ijill rtr+ drlJ$" U pl,aii-,| al=llJYl gtr.Ltii .i" cie$ll e$'lll qJ&. LJs^ll 4s^ .,.i is.3*all

o-pi

g,.o.r-,!l .!rc

JLr

I I I I I t I I I I

t rsHl ep iuJilt

ol-lgis.r

[cJiill 4*,Jia

/a;r.

ir.;.r

d+J l,..Sa1

e-+

I

I T

t I 4-,$6lt

4.JS

;+ - >jJl r.rc ,-rll ll i'-"1-: . 1Y. \ jgL _ -ol tYl .rJ3Yt 6rt-;

I I I

I t

FJll iFJllritl

" g3,i,c!.cll3

(t.o

d.y-rf

r1,13:lti-r-r-)

fle

e.*l

Al 6J#"i lJf.el 1$3 ": ,Jlri Al jl3

fi**llAl drr

d''l.J;ij.ll rtr+ d',|#i pL,a$-'l jue drl..lLiYl crt .bt$ ,g.ril! e$lll "b 4j3.:ll L_f"lt 4s" .,.l

,* \ull

o-;*,i g..r..!l

eilrJt ep

rp $ti

iLJill ol.r5r is.-,1.r jil e& [dill 4-,riA /i;u e*,Sl]

,,Je ipli"ll sL^rl .,11.;ll .,;.s ,r{r ..t Jldt i" -l++ .Lro ..r .i

i.lt*,,.rrtt

4."tiCl lJlS

:-rlt +t dlJl a.-B a;: .*^,Jl i-u!l iS.L^ll - ;rpY. \ . 3;L - -r\ tY\ t1)l 6:l-;

,o.-i

ervi:Jl

cjll. c,lg L.#l

clllniJf tF+

I I cilJrrijl rL,.aii*l

cJJl*l'Yl crlr.Lli: $e #$ll LrS.lt 45" ,i qjj.hll 0.,,1r*i

g*r.,/l

rp

T

t I

rJt3

Uali'Ll.oll

T

L:-!-,,-l all; !:.l)-l--i e Y. . . eb d a-Jsylt .-!JUl a*.., #r +_r_! i i.--,j'ill i-",il. I . ;rF ,Js.9 (5lEll-,r ,dtir:ll qFltj.jl C+iiI ,J!. Jr lrlr:.r.rll .[JJl .aJ1LI 1^$ll f +,,Eii ) Clr C!t^6 9i ric. ,gslJ C+.lill Jl-".I r u ll4l .' ll c.il-!.-ll ,-;"+ 3 .!i"rilJ 6+i,il| cJu gl-.=il-l J! ilt-,jl ora ,-ir6: dJJLi)l crtJl& .a;r,*^.Jl 4J+!l is.L.ll ,LJS^ll oS. i+r" e ^l*lt

t I I I I I I

Jn"

J gll3.rJl :LJ. d$ al-.lJ| d!\Ul rL-i:-l p cJ*-oi r-jc al-jl$ e-+ i ,JJl 4..,p ri,a;iiJl aS>X a.-.-i-ll diJlJl L#-i i _s+ l-u.lS el \r",.,L C;S-FJI iiti"ll 6-tli 6rr U. ro 6lJ1 il slJl.i: :j,.c +>t & dF Jt6ll ,:U1* cF)-i3-l i .:*,JJr.ll l:lJli3ll cjib. c.l. !-.6r i+'1rl (eu$-l :;+ J. rr) .+* Jir*FS e-Uj els3-1+ -r;,:nl drl;u.lS cl+,,'l .u-e ci,-!&.Il

c:l+S-Fll

.,

:1.:.11

"o-,11-rl

.4+B

r . /\

i-+;.- .:lJL!,) dillttj i* i!)'ll .:llS-r dpr:ll e+il1 jr'-3 6113ll :*^,, dr# LU.ll ejlill lra 6:& .d,sr:! e+idl d&-J

sl,,
,i-'-tl3 ' r. ,!i:3ll ,JtLll C+bll cJ:'- dp riiul drl-,lJil eJ$ ej,.ll o-J,n-,j ,rt+.+-":l Y. .

a;-t-t-rll

8)tJl r '-: lX ,illJl u" aibi- cjl

. ed +-.,S,r-pYl 6Jtll ;4^,J &lel;r;)l

os

4i:.-ll !:lJLi,)l

.ljll'3.il Jr'-ll etit c-pJl gp.:Q

Llb d ciJtlt Cy crL+JrJl ,C.Llj:ll 6l-,rL- ,r.,;c :dF JSI iL'.I..:ll cJ-lJl i'ai rt--r.) ellr-e rp .ClEill nlj.,* so r,-!lJl J OlJrJl ,il,ri:ll ,J.-jl=]]l e11-.j sl3 &r:ll e+:,:ll ,J:*. :$ crJil rg:lt u".Bl 'j+lJ LFJ,n-t ,,,l--ao)l iS.,,.j'r .LeEL-Yl sj1 -Yl gisJYl aiJ'3 ll 4ir.tJl 4-,lJil W-l'.! ri ':l.,llll o',,'!Ul e \+ll LFJ'6-;

.j

a:,=vl

..if!l aSL'll cJ3r d 6;i crLl-.p 31Jl 'j;Jl u-J,s-,..l L;i ea;r,'-,Jl $r i,.-il l.ra o11 rj.lc .rlc ,J l-,1.r,qlS 6;.rtuj ,',;in;.JirS U,*lAll o -r+ll cj:l+ l+ . L,itll dlFl .-l1'J'"1 Ji'+ Ji' J"J"-: ,,Jc l.!'.-'! 6lti:lt clir' ';;;iJl ;-,11i)l C--"1^+ c!+; I'tili cJ"rUl c i+ll dJl'r'j"''-jll ,/ 4+1J!,oll d,EtJill -r L-:S.lt aL a.)o +l;i'+ J#l-.-ll c jaJJ li: u,aJ.luJ f: ,aJ6. J;ll 6lJt :,)l citJti: e \uy drL J'-. 4-5+ -r-r--r Cfw ..lS ir. ..," ;+ J! a-AL:)t+ ri-,|!l drlJlii Jl-ii" Y e--Ell lJeJiJsJ elrii-ti dlliJ drul+Jl a.ll--l Lsi,".ll o+Lll cJ.U eJFll iJ *2n", .41 4rr. ,r-lo &.-.ll d-J g. -rUJ f . ti ;.slil rSBll i;-*., lJir-rFsll d e$-r e-Li:

1;rll e i.= &.Bll ej+1. I O--:+.a-ifri^I d'll$J^ll 0+ Jltiill ets uF d ;l1st a-lr 6!-,yt-..,- lc ci'l:6i!l dJJSJrlt eirliJl ,rJt gl-.,y3rJt :g. dS 4-,lJJ U;J.-e r4-tl_lil .di15l Ji,.ill n!-i.s- C-Ell e-,l,+!l , '-il: l:l5i3.6lJlJl (o-,ult dir:ll C+tll du.,J:L:J elil dl,-ill 8lJ1 .' ^ll s-E {lL+JFll JA16 ,JC-1 ,} l'.ljyt cj;Jl I..,-: ,rl;J a5g ;t''* I

.:t-,,I-,pt! sit,a:Yl &Lll" I &tiJ L^tJi r ;j',U.ll ,/ ir" .+;-Ji:,".11 ^Slll Jir:ll;t!l i+iill cJr'- ;p$ 0'\t :,:--+ tl dL:ll '-r+ll d a$.'i-ll e-,r)l

I I

I I I I t t I t I I t t

I I I I t t

T

lJ-t

i:ild:is,. eJEjl

e

j+ll ,,'-i!,4--.-ll ::jih;^ll nElJjll ,rJo frt&l

4*,,lJAll

!*l

.4-jl:r jJ*ir JL-ill

(t-l+t :tt di)-l-.'ll J;iE) L,""tiJ J.,-.iI LJ3!l e+ ,'t- ,el;i il-' C--:il 'j+ll ..1$-ill ,rb q1.*lll JJ ,j$l !F elj,r.ll

I ^rl cLl.-.il| dr!-.:-- i1c Li6. rj+-., o-'ltt ci)rl--ll g,.a,:r,+X et! Jl ,,j_,rdl :e. jsl lLl":ll d.il-tJl J*ril dllr iiL:. 4i't- crl-l.;r ,J kjl \r',.,1 c}31-i::

.drlJtJl u-Jo ,eltBll

drl-.,1L"..rrc alJri:ll ciUiJ^ll

., ikll

i-;6.,.-ll crlJL,,-.ll j JtlSJl c,bh 4;jj;p ).-i" $-: s"rlJl oj+!l cj:14 ,.!g$l e+ill ,Jr*. a.+ 6^ c$lr: ArJ^ d J.-C J1 lll J,! ,-ili3Jyl 6lJL,,4r .4i" h;.ll riEl++ll C.+ g,. eljll

(.+rt!l eirrill :.'^- gr,ii*',tt) u"JJl d.-ill ,o;;+:tl erJ.ill ls,a gr r$.:lt 1J.:,LrLll :JtXl jstl .,Je'l;J a'r.-j ;.^:q

J,.-ill ue cJ.rl^r qf.6+J ,ai!:i- .:t++d. irc plr.i-l; fu-1liJl a-:-,r-r:^ll c,L"l'-lt 4+r.i (rJSir-) q-tsA;lslJl e-u-';', d#!n Lll.-ill t.u 6-,14 .4-,lJJ1 .,r 4J--ll JIF)I C. lielrill .

(a-JJJI CjBj i.i^Al)

i/\r'"o: -t:-rll #ly -.g......l.LlJJl .J. tCl l* 'o-j-i,)l fiF--i e-tli:ll aJ..-: ,J, AIEII ir:"ll Ji. illl

elJl

Ljl,-ill q^ni 3.'r..].-itl lrA .tstl J! a-it;Yl+ ,a+-!.ill e-l-x

]+li1ll,i.;i.:.llrl,,n .^:ll

(ot+.-Ct, dil-1Ej:-,)l) g.t3ll Ll.-ill

I :i ,ol'. 131-,Yl u,a)i3,dl i ,U4.1J13 eljl dl.-ilL f.iililll e5l3il slc' Irt+ d+-Jl lra d iil.G6,., ll sl.,lJrll dli3 &1--. Ei:-)l .JA .,le ol;.,a3il| '':tr ,

I I T

I I I I T

I

I t I I T

I I I I

I

I t I I I I I I I

AN INVESTIGATION OF THE EFFECTS OF SOME TRAFFIC PARAMETERS ON THE SATURATION FLOW RATE AT SIGNALIZED INTERSE CTIONS IN

MAKKAH ALMUKARRAMAH

By Khalid Abdulrahman Osra

T

I T

I I I I I I t I T

A thesis submitted for the requirements for the degree ofDoctor ofPhilosophy in science ICivil Engineering/ Transportation Engineering]

Faculty of Engineering King AbdulAziz University, Jeddah Jumada'1, 1431H - May,2010G

I

t I I I I I I I

i-..

FFJI i,"rJl

.:,tt

e*r

T

I I I I I I I I

I

I I

"

(l.o

Ogi"3"lll4l.*,lf fSl"e 41 rr.Xlg l.tLcl 1p3": ltr: At jl3 fi#dtdll d&. ri.l.3ill6.13-)

AN II\"VESTIGATION OF THE EF'FECTS OF SOME TRAF'FIC PARAMETERS ON THE SATURATION FLOW RATE AT SIGNALTZED INTERSECTIONS IN

MAKKAH ALMUKARRAMAH

By Khalid Abdulrahman Osra

A thesis submitted for the requirements of the degree of Doctor of Philosophy [Civil Engineering/ Transportation Engineeringl

Supervised By

Dr. Hamid OmarAlbar Prof. Dr. Mohammad Jobair Bin Alam

FACULTY OF ENGINEERING KING ABDULAZZ I.JNIVERSITY JEDDAH _ SAUDI ARABIA Jumada' I, 143lH - May, 2010c

I

l t I t

AII INVESTIGATION OF THE EF'FECTS OF SOME TRAFFIC PARAMETERS ON

THE SATURATION FLOW RATE AT SIGNALTZED INTERSECTIONS IN MAKKAH ALMUKARRAMAH By Khalid Abdulrahman Osra

This thesis has been approved and accepted in partial fullillment ofthe requirements for the degree of Doctor of Philosophy [Civil Engineering/ Transportation Engineeringl

EXAMINATION COMMITTEE Internal Examiner Extemal Examiner CoAdvisor

Advisor

Name Sabry Shihata Hasan

Al-Ahmadi Md. Jobair Bin Alam Hamid Albar

Rank

Field

Professor

Transportation

Associate Professor

Transportatiorl

Professor

Transportation

Associate Professor

Transoortation

KING ABDULAZZ LINTVERSITY Jumada I, 143lH - May, 2010G

v

Signature

'4--t

1'V

*Y+>

Dedicated to

I wish to acknowledge my parents, my wife, my sons, brother, sisters for their constant support and encouragement which made this thesis possible.

ACKNOWLEDGMENTS

The author wishes to express his profound gratitude and sinceiest appreciation to his advisors Associate Prof. Hamed AI-Bar and Prof. Jobair Bin Alam for their invaluable guidance, encouragement and support throughout the study. The author expresses deepest gratitude to His Excellency Dr. Osama Albar, the Mayor of Municipality of Makkah, for his moral support, instructions and encouagement that gave me the strength to complete this study. Sincere gratitude is extended to the Dean of the College of Engineering and the Chairman of the Civil Engineering Departrnent for their thoughtflrl advice, support and encouragement, the same goes for all members and friends in the laboratory for their assistance.

The author expresses his special gratitude to Eng. Mohammad Bahareth, the General Director of the Lighting Management in the municipality of Makkah region, for his encouragement and unlimited support.

The author expresses his gratitude to King Abdul Aziz City for Science and Technology (KACST) for their financial support (Research Number. GSP 16-l l4). The author expresses a special thanks to Traffrc Police Departrnent especially to Eng. Shareef Saleem and Eng. Yahia Al-Husainy for their continuous assistance and supplying support accessories and materials.

I T

I T

i 1V

T

l T

I I

AN IIIVESTIGATION OF THE EFFECTS OF SOME TRAFFIC PARAMETERS ON THE SATURATION FLOW RATE AT SIGNALIZED INTERSECTIONS IN

MAKKAII ALMUKARRAMAH

t

Khalid Abdulrahman Osra

Abstract

T

t I I I I T

I I I T

I T

I

i t

Highway Capacrty Manual (HCM 2000) provides adjustment factors for addressing the impacts of operational factors on the standard value of saturation flow rate. Thus, an effective value of the saturation flow rate can be assessed at any intersection that does not meet the ideal condition. This dissertation mainly aims to investigate the saturation flow rate and estimate correspondingly effect of some tmfific parameters on saturation flow rate at signalized htersections in Makkah, Saudi Arabia. The adjustment factors for Utums, heavy vehicles, number of tfuough lanes, and lane width were investigated. Data was collected at five non CBD signalized intersections using video cameras that mount on the nearest light pole of the studied approaches. The headway data were extracted from video tapes using video-editing software with an accuracy of 1/30 sec.

Headways were grouped to different categories based on both vehicle types (passenger cars or heavy vehicles) and type of tuming movement (left- or Uturning vehicles). Total duration of the study was 18.5 hours. Sample size includes 1589 and 1639 successive headways of U-turn and heavy vehicles studies respectively. Valid cycles were included in the analysis for the four selected studies, which include U-turns, heavy vehicles, number of tbrough lanes, and lane width. They were 234,327, 564, and 560 cycles respectively. In the total, 1015 left-tum vehicles and 574 U-tum vehicles were obtained. Also, 1277 passenger cars and 362 heavy vehicles were obtained.

In the analysis of left-tuming vehicle preceded by a left-turning vehicle in a lefttum lane, the average headway is found to be 1.90 seconds. When a U-tuming vehicle preceded by a U-tuming vehicle in a left-tum lane, the average headways is found to be 2.37 seconds. Analysis reveals that the larger the percentage of Utuming vehicles, the greater the effects of U-tums on the saturation flow rates. When a passenger car preceded by a passenger car in a through lane, the average headways is found to be 1.54 seconds. When a heavy vehicle preceded by a heavy vehicle in a through lane, the average headways is found to be 3.01 seconds. In addition, the study implies that the larger the percentage of heary vehicles in the through lanes, the greater the effect of heavy vehicles on the saturation flow rates. Based on headway distributions for different cases of vehicular movements corresponding adjustment factors are estimated at different percentages of U-tums

I I

I T

I I t I t I t I I I

and heary vehicles. The results show that the adjustment factors of both U-tuming vehicles and heary vehicles are similar to other studies in the USA.

The adjustnent factors for number of lanes for approaches having through lanes of 1,2, and 3 are found to be 0.86, 0.93, and 0.95 respectively. The adjustment factors for lane widths with 3.3 m, 3.5 m, and 3.6 m are found to be 0.84, 0.97, and 1.0 respectively. The results indicate that the overall saturation flow rate increases with an increasing number of through lanes and lane width in the through lane group. In addition, the adjustment factors of number of lanes and lane width in the through lane group is low compared to those developed in the USA. The average headways estimated for all the four studied parameters, are low compared to those obtained in USA. It implies that the driver behavior in Saudi Arabia is more aggressive as compared to other countries. Also, the ideal saturation flow mte in Makkah (2500 pcphpl) is higher than that was suggested in HCM 2000 (1900 pcphpl). Investigating the driver's headway choice behavior it was observed that the mode of the headway distibution, rather than the mean, is a better representation of the central tendency particularly for the t}rough lanes. The methodological approach presented in the thesis can be utilized to investigate other factors, which include axea type based on land use, area population, rightturns, left-turns, and weather. It will also help in the preparation of Saudi Arabian Highway Capacity manual. The study outcomes can be utilized in the planning and design of traffrc signals and geometric features of the roadways. The findings of the study will also be useful for the traffic simulation models to analyze microscopic characteristics ofthe traffic flow in Saudi Arabian conditions.

T

I T

I I

i I

vl

T

T

I T

I I I I

TABLE OF CONTENTS

ii iii

Examination Conmittee Dedication Acknowledgment Abstract Table of Contents List of Figures List of Tables List of Symbols and Terminologr

iv

X

xii xiv

Chapter I: Introduction

T

I I I I I I

I I I

t

i t

1.1 Problem Statement

I

1.2 Research Objectives 1.3 Research Organization

A

Chapter II: Literature Review 2.1 lntroduction 2.2 Survey of SFR Related Intersection Capacity 2.2.1 Saturation Flow Rate Definition 2.2.2 Relationship between Intersection Capacity and SFR 2.2.3 Relationship between SFR and Headway 2.3 Review of the Base Saturation Flow Value Estimation 2.4 HCM 2000 Method for Estimating SFR and its Adjustment Factors 2.4.1 Lane Width Adjustment Factor, fiv 2.4.2 Heavy Y ehicle Adjustment Factor, f11y 2.4.3 U-tum Adjushnent Factor, fu 2.4.4 Number of Through Lanes Factor, fi 2.4.5 Other Adjustrnent Factors 2.5 Other Methods for Estimating Saturation Flow Rate 2.5.1 Regression Analysis Models 2.5.2 ANN (Artificial Neural Network) Models 2.6 Previous Studies Related to SFR in Saudi Arabia 2.7 Studies on Capacity Analysis of Signalized Intersections 2.8 Summary

Chapter III: Data Collection and Analysis 3.1 Introduction 3.1.1 Headway Method 3.1 .2 Regression Technique Method 3.1.3 TRL Method 3.2 Study Methodology

vll

8

9 9

ll

15

2I

a^ 25 30 .)J

39 43 45 45

46 47 49 50

5t 51

52

52 54

I

I I t I I t T

t I I I I I I I I I I

i T

Data Collection Methodology Time Schedule of Data Collection TrafEc Condition Durins Data Collection Camera Position Selection of Study Area 3.7.1 Background 3 .7 .2 Citet'ra for Selecting Case Study Intersections 3.7.3 Preliminary Site Assessment 3.7.4 Selection of Case Studv Intersections 3.8 Data Processing 3.8.1 Background 3.8.2 Data Processing Technique 3.9 Sample Size 3.9. I U-tum Adjustment Factor 3.9.2 Heavy Vehicle Adjustrnent Factor 3.9.3 Adjustment Factor for Number of Lanes 3 .9 .4 Lane Width Adj ustment Factor 3.10 Summary 3.3 3.4 3.5 3.6 3.7

Chapter IV: Analysis of Basic Saturation Flow Rate (SFR) 4.1 Introduction 4.2 Headw ay Distributions Analysis 4.2.1 Assessment of U-tum Adjustment Factor f!/ 4.2.2 Assessment of Heavy Vehicle Adjustm ent Factor 6uu) 4.2.3 Assessment of Number of Lanes Adjustrnent Factor (fy) 4.2.4 Assessment of Lane Width AdjustmentFactor (f,) 4.3 Estimation of the Ideal Saturation Flow Rate 4.4 Summary

Chapter V: Estimation of Adjustment Factors 5.1 Introduction 5.2 Analysis of U-turn AdjustmentFactor (fu) 5.3 Analysis of Heavy Vehicle Adjustm ent F actor (nr) 5.4 Analysis of Number of Lanes Adjustment Factor (fy) 5.5 Analysis of Lane Width Adjustrnent Factor (f,/ 5.6 Correlation between Observed and Expected SFR 5.7 Summary Chapter VI: Empirical Validation of Model 6.1 Introduction 6.2 Model Validation Plan 6.3 Comparison Among the Results 6.4 Summary Chapter VII: Policy Implication of the Study 7.1 Intooduction 7.2 Implications of Findings 7.3 Expected Effect on Planning 7.4 Expected Effect on Intersection Design

55

56 57 58 62 A' 63 64 65 79 79 84 87 88 90

92 93 94

95 95

96 98 99 101

103

104

105 105

108 110

112

t14 .T

JJ

t34 t34 135

139

140 141

142

t43

I T

t T

I I I

7.5 Expected Effect on Signal Design 7.6 Expected Effect on Traffic Simulation 7.7 Summary

Chapter VIII: Conclusion and Recommendations 8.1 Conclusion 8.2 Major Findings of the Study 8.3 Summary of the Implication of Results 8.4 Recommendations for Further Study

List of References

144

t44 r45

146 147 150 151

t52

I I I I I I I I T

I I t I I I T

I I I I T

LIST OF FIGI]RES

Figure '

1.1 | .2 2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.9 2.I 0

Discharge Flow Rate Characteristics at Signalized Intersections Framework of the Research Study Measwed Saturation Flow Rate (Teply 1995) Attributes of Traffic Flow at Signalized Intersection (HCM 2000) Traffic Flow During the Green Period from a Saturated Approach Headway Determinations (Traffic Control Handbook 2005) Traffic Intemrption at Signalized Intersection Approach Saturated Headway Time and Starting Lost Time (Lee 1995) Comparison of Headway Values in Dhaka and Yokohama Location of Queued Vehicle in Dhaka and Yokohama Departure Headway at Signalized Intersection in Different Studies List of U-tum Adjustment Factors Observed in Three Studies 2.1I Comparison of Average Discharge Headway (Liu 2009) Video Cameras Mounted on the Steel Stand Steel Stand Mounted on the Light Pole Video Cameras and Steel Stand Mounted on the Light Pole Selected Light Pole Mount of Al-Kakya Intersection Cable connection between the LCD Screen and the Video Camera Position of Video Camera Relative to the Traffic Flow Direction Real Image of the Video Camera from Its Mounted Position Location of the Preliminary Selected Intersections in Makkah Location ofthe Final Selected Signalized lntersections in Makkah 3. Google Image of Badr Intersection North-Bound Approach of Badr Intersection 3.1 3.12 Layout ofthe (NB) Approach ofBadr Intersection 3.13 Google Image of Um Al-Joad lntersection 3.14 West-Bound Approach of Um Al-Joad Intersection 3.15 Layout of thb (WB) Approach of Um Al-Joad Intersection 3.16 Google Image ofAl-Beban Intersection South-Bound Approach of Al-Beban Intersection 3. I 3.18 WestBound Approach of Al-Beban Intersection 3. 19 Layout of the (SB) Approach of Al-Beban Intersection 3.20 Layout of the (WB) Approach of Al-Beban Intersection 3.21 Google Image of Al-Kakya Intersection 3.22 North-Bound Approach of Al-Kakya Intersection 3.23 South-Bound Approach of Al-Kakya Intersection 3.24 Layout ofthe (NB) Approach ofAl-Kakya Intersection 3.25 Layout of the (SB) Approach of Al-Kakya Intersection

3.1 1.2 3.3 3.4 3.5 3.6 3.7 3.8 3.9 l0 I

7

Page 2

7 10 13

l.t 15

l6 t7 t9 t9 20 36 39 58 59 59 59 60

6I

6l 65 66

67

o/ 68 68

69 69 70

7l 7l 72 72 73

74 74

t) 75

I I T

I I I I I I I I I I I I I I I I I I

3.26 3.27 3.28 3.29 3.30 3.31 3.32 4.1 5.1

5.2 5.3

5.4 5.5 5.6 5.7 5.8 5.9 5.10 5.11 5.

12

5.13 5.14 5.15 5.16 5.17 5.

18

5.19 5.20

6.1 6.2 6.3

Intersection Intersection Intersection

Google Image of Al-Khaldya West-Bound Approach of Al-Khaldya Layout of the (WB) Approach of Al-I(haldya Hybrid TV Connection Process ofthe TV Headway Classification Used in the U-ium Headway Classification Used in the FIV Average Headway of Vehicles in Through Comparison of U-Tum Adjustment Factor with that of Other Studies Comparison of Heavy Vehicle Adjustment Factor with that of Studies Comparison of Number of Lanes Adjushnent Factor with that Other Studies Comparison of Lane Width with that of Other Studies Histogram of Headway Category frpp in Through Lanes Histogram of Headway Category ftps in Through Lanes Histogram of Headway Category fr7p in Through Lanes Histogram of Headway Category fts7 in Tlrough Lanes Cumulative Headway Distribution of Category ftpp in Tlrougb Lanes Cumulative Headway Distribution of C ategory hppin Through Lanes Cumulative Headway Distribution of C ategory hpp in Through Lanes Cumulative Headway Distribution of Category ftss in Through Lanes Histogram of Headway Category frz1 in Left-Turn Lanes Histogram of Headway Category /l1u in Left-Turn Lanes Histogram of Headway Category y'ruz in Left-Tum Lanes Histogram of Headway Category huuin Left-Tum Lanes Cumulative Headway Distribution of Category ft4 in kft-Tum Lanes Cumulative Headway Distribution of C ategory h1u in Lanes Cumulative Headway Distribution of Category hut in Lanes Cumulative Headway Distribution of Category huuin Lanes Geometry of Al-Beban Intersection Appeared in Synchro Lane Window of Al-Beban Intersection Appeared in Synchro SimTraffic Animation in Synchro

Card

Card

76 77 77 84 85

Study Study lanes

Other

of

89 91

96 108 1

I

0

ll2 114

t20 120 121

t21 122

lz3 123

t24 128 128 129 129 130

Left-Tum

1

Left-Tum

131

Left-Tum

132

3

I

138 138

t39

I I I I I

I I I I t I I I I I I I t I I t

LIST OF TABLES

Table

Page

2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.9 2.1 0

Departure Headways (seconds) as Observed in Beijing Previous Studies' Saturation Flow Rates Saturation Flow Rates (vphpl) in Stellenbosch, South Africa Lane Width Adjustment Factor of SFR Based on HCM 2000 Effect of Lane Width on SFR (Agent and Crabtree 1983) Effect of Lane Width on SFR (Zegeer 1986) Adjustment Factor of Heavy Vehicles Proposed by HCM 2000 Average Headway Proposed by Tsao and Chu (1995) Three Movements' Saturation Headway Comparison (Liu 2009) Saturation Flow Rates Relative to Condition (rphpl) 2.11 Saturation Headways and Lost Times for Through Movement 2.12 Number of Valid Cycles and Headway (Le et al. 2000) 2.13 Saturation Headways Observed by Le et al. (2000) Time Schedule of Data Collection Geometric Characteristics of Study Sites Sample Calculation of Time Headway Presented in Excel Sheet Sample Size of the U-tum Adjushnent Factor Sample Size of the Heavy Vehicle Adjustment Factor Sample Size of the Number of Lanes Adjustment Factor Sample Size of the Lane Width Adjushnent Factor Average Headways and Results of the COV Test for U-turn Adjustrnent Factor Results of the ANOVA Test for U-tum Adjustment Factor Average Headways and Results of the COV Test for Heavy Vehicle Adiustment Factor 4.4 Results of the ANOVA Test for Heavy Vehicle Adjustment Factor 4.5 Average Headways and Results of the COV Test for Number of Lanes Adjustment Factor 4.6 Results of the ANOVA Test for Number of Lanes Adjustrnent Factor 4.7 Average Headways and Results of the COV Test for Lane Width Adjustment Factor 4.8 Results of the ANOVA Test for Lane Width Adjustment Factor 4.9 Descriptive Statistics of Headway for Ideal Saturation Flow Rate 5.1 Adjustnent Factors for U-Tums 5.2 Adjustment Factors for Heavy Vehicles 5.3 Adjustrnent Factors for Number of Lanes Adiustment Factors for Lane Width

3.I 3.2 3.3 3.4 3.5 3.6 3.7 4.I

4.2 4.3

xll

2l z) 24 zo 27 28

)z JZ 38

4l 42 44 44 57 79 87 89 91

93

94 97 98 99 99 100 101

102 102 103

107 109

lll

113

5.5 5.6 5.7 5.8 5.9 5.

10

6.1

Adjushnent Factors ofthe Studied Intersection Approaches Observed and Expected SFR in Through Lanes (Model l) Headways of Different Categories in Through Lanes Observed and Expected SFR in Through Lanes (Model 2) Comparison between Observed and Expected SFR in Left-Tum Lanes Headways of Different Categories in Left-Turn Lanes Comparison of Total Delay and LOS in Different Approaches

xllt

115

tt7 t1i

t25 127

r32 tJ I

T

t I

l I I I I T

I I I I I I t t I I I I

LIST OF SYMBOLS AND TERMINOLOGY

AR

ARRB

Br BS Bo,

c CBD CCG

cov EB Eg

rbb

f IHV

f, ILT Ip rRT ru

ru

FOV G o

G G" h

All-Red Time, seconds. Australian Road Research Board. Regtession Coefficients. Business Area. Capacity of Approach, vph. Cycle Length, seconds. Central Business District. Canadian Capacity Guide. Coeffrcient of Variation Statistical Analysis. East-Bound. Passenger Car Equivalent for Healy Vehicles. Random Error Term. Adjustrnent Factor for Area T1pe. Adjustment Factor for Local Bus Blockage. Adjustrnent Factor for Grade. Adjustrnent Factor of Heavy Vehicles. Adjustment Factors for Non-Ideal Conditions. Adjustment Factor for Left-Tums. Adjustment Factor for Parking. Adjustment factor for fught-Tums.

Saturation flow Adjustment factor for U-tums in an Exclusive Left-Turn Lane with Protected Phasing. Adjustment Factor for U-tuming Vehicles. Field of View Green Time, seconds. Effective Green Time, seconds. Gradient, (%). Effective Green Time. seconds. Average Queue Discharge Time for Each Tuming Vehicle, seconds.

HCM HV h,

n

N NB

Nr* D r

auo

Pt*

Highway Capacity Manual Heavy Vehicle

Average Time of Headway between the Departure Vehicles During the Saturation Discharge Condition, seconds. Number of vehicles in the Queue, seconds. Number of Lanes in the Lane Group. North-Bound. Maximum Number of Suweyed Vehicle Per Cycle. Percentage ofAutos. Percentage ofBuses.

xlv

I

t I I I I I I I I

I t t I

I t I

pc/Wln

P"^

pcphpl

PC PCE PCU P.t D rff D Put

RC RD S S

SA SB Sccc SFR SH SHcv Si

SL so T1

tr

T. T"

tcu/tr TL U USA VIVDS vph vphgpl vphpl WB

Y

Passenger Cars Per Hour Per Lane. Percentage of Cars. Passenger Cars Per Hour Per Lane. Passenger Car Passenger Car Equivalent. Passenger Car Units. Percentage of Right-Tuming Vehicles. Percentage of Trucks. Percentage of Two Wheelers. Percentage of U-tuming Vehicles in the

kft-Turn Lane. Recreational Area. Residential Area. Saturation Flow Rate. Prevailing Saturation Flow Rate in the Lane Group, vphg. Saudi Arabia. South-Bound. SFR calculated using CCG, pcphpl. Saturation Flow Rate, pcphpl. Shopping Area" Table 2.1l. SFR Value Estimated Conesponding to HCM, pcphpl. Ideal Saturation Flow Rate, pcphpl Speed limit (60 or 80km/h). Ideal Saturation Flow Rate Per Lane, (=1900 pcphgpl). Start-Up Lost (Delay) Time, seconds. First Headway in the Queue, seconds. Clearance Lost (Delay) Time, seconds. Second Headway in the Queue, seconds. Elapsed Time Since the Start of Green from the Vehicles Stopped Position z; to Enter the Intersection. Through-Car Units Per Hour. Number of Through Lanes. Average U-turn Percentage in the Exclusive Lefttum lane. United States of America Video Imaging Vehicle Detection Systems Vehicle Per hour Vehicle Per Hour Green Per Lane. Vehicles Per Hour Per Lane. West-Bound. Yellow Time, seconds.

T

I

I xv

T

I

l I t I I I I

Chapter

I

Introduction

1.1 Problem Statement

Signalized intersection is the most common form of traffic control measure used in

T

the urban areas of both developed and developing countries. Saturation Flow Rate

I I

maximum queue discharge mte

T

intersection approach. The discharge rate reaches

I

satuation flow rate) after a short time (i.e., four or five vehicles) from the onset of

T

I I I t I I I

(SFR) is one

of the core entities in

design and operations

intersection. Conventional theory suggests that saturation

of traffic

of the signalized

flow is the steady

across the stop-line

of a

signalized

its maximum value

(i.e.,

the geen time and then remains constant during the green time period until the queue is completely dissolved, as illustrated in Figure 1.1. In this figure, the slope

of the straight part of the cumulative curve of departed vehicles

represents the

value of saturation flow rate (S). It shows also two types ofdelay time occur at the signalized intersections; start-up delay (T1) and clearance delay (T2). The start-up delay occurs during the onset of the green time due to the reaction time of drivers and from the recovering process of speed from the stop condition to the desired speed

of drivers. This delay usually takes place starting from the first four or five

vehicles in the queue. On the other hand, the clearance time occurs from the onset

I

t I I I t

of the yellow time until the onset of the green time of the next phase. Minimizing these two types

of delay is one of the major tasks of traffic

researchers and

engineers since it has a direct impact on the intersection capacity.

6 ,a

T

t T

I I I I I I I I

I t t I

Curenl phase Next phase

c: green.

Y: yellow, AR: all-red, S: saturation flow rate,

C.: effective green iime, 1,/: start-up lost time 12: clearance losl time

N-o: maximum

number ofsurveyed vehicle per cycle

Figure 1.1 Discharge Flow Rate Characteristics at Signalized Intersections

Saturation flow rate is defined in Highway Capacity Manual (HCM 2000) as "the

equivalent hourly rate at which previously queued vehicles can traverse an intersection approach under prevailing conditions, assuming that the green signal is

available at all times and no lost time is experienced". In (HCM 2000) manual, a standard value of saturation flow rate

of 1900 pcuphpl is estimated

based on data

collected at signalized intersections in USA in an ideal condition. In addition, it is stated that this standard value is influenced by various extemal factors. These

extemal factors are classified into three categories: l) factors belonging to traffic parameters such as percentage of heavy vehicles, and percentage of left, right, and

I T

I I t t I I I I I t I I I I I t I T

t

U-tuming movements,

2)

factors belonging

to control

mechanisms, such as

protected or permitted left-tum, and protected or permitted pedestrian crossing, and

3) factors belonging to geomehic design of intersections such as lane width, number

of

lanes

of each movement, lane configuration, grade, lane

clearance,

parking permission, and bus stop locations.

HCM 2000 provides adjustment factors for addressing the impacts of operational factors on the standard value

of saturation flow rate. Thus, an effective value of

saturation flow rate can be assessed at any intersection that does not meet the ideal

condition. However, the proposed value of standard saturation flow rate and its adjustrnent factors can not be applied as documents, since the characteristics

a

default value as found

in

of vehicles, drivers, and intersection

USA

design

vary from country to country and also from city to city in the same country.

Traffic software and programs are drastically increasing during the last decade. ln these programs, the standard satuation

flow rate and its adjustment factors

are

intensively used as input parameters. Trafftc engineers would often use default

values from abroad software developers because the lack researches, especially

of local data and

in the developing countries. This can lead to wrong results

and inappropriate decisions based thereon. Accordingly, several researches were carried out in order to estimate the traffic parameters such as saturation flow rate and its adjustment factors that match the local conditions over a number

ofcities.

Despite the importance for determining the standard value of saturation flow rate and the influence of the operational factors under the local conditions, no serious

study tackled this issue at signalized intersections in Makkah, Saudi Arabia (SA). Thus, this study tries to cover this research point as a step towards implementing

conorehensive rcsearches

r: other citics in SA and to cover more factors at

I

I I t t I I I I I I I I t I I I

I I I I

intersections

of

case studies.

It is therefore

estimated that the output

of

this

dissertation will remodel and reassess the default values of saturation flow rate and

its

adjustments

of

some operational factors that can be applied at signalized

intersections under the local conditions of Makkah.

1.2 Research Objectives

This research mainly aims to assess saturation flow rate and some adjustrnent factors

at Makkah in Saudi Arabia. To

associated

achieve this targe! this

dissertation is aimed at following objectives:

o

To estimate the standard or ideal value of saturation flow rate at sisnalized

intersection approaches in Makkah.

o .

To determine adjustrnent factors for the effects U-tums, heavy

number

of

lanes, and lane width on the saturation flow rate

at

vehicles,

signalized

intersection approaches in Makkah.

o

To

assess and characterize the differences between the estimated value

saturation

of

flow rate and the selected adjustment factors at Makkah and the

conesponding values in other cities over the world.

1.3 Research

Organization

The framework ofthe research is shown in Figure 1.2.

This dissertation consists of 8 chapters including the introductory chapter. The remaining chapters (ch 2-8) are organized as follows: Chapter 2 provides a literature review which documents some previous studies on

the topics analyzed

in this dissertation,

such as estimation methodologies of

I T

I

satuation flow rates at signalized intersections, factors that affect the standard saturation flow rate, and the correction values ofthese operational factors.

T

Chapter 3 presents a background of the total signalized intersections in Makkah. It

I I t I I I I I I t I I

also identifies the selected study sites included in this research. In addition, it describes the required data for achieving the objectives of this research as well as

the applied methodology for collecting this data.

t t I t

also presents the criteria for

selecting case study intersections for data collection purposes of the research work presented in this dissertation. Then,

it

describes the data processing technique. It

also presents the sample size of the different studies for the selected adjustment factors that impact on the saturation flow rate.

Chapter 4 provides an analysis on the saturation flow rate at the selected study intersections under the ideal conditions and then investigates the standard value

of

saturation flow rate in Makkah intersections.

Chapter

5

presents an analysis

of the impacts of

some haffrc parameters on

saturation flow rate in order to develop adjustment factor models. Four parameters are investigated in this chapter; which include percentage of heary vehicles, lane

width, left and U-turn movements. The correlation between observed and estimated

sahration flow rate (SFR)

is

included

to signi$ the effects of the

studied

parameters on SFR. Then, a detailed headway distribution analysis is performed in

both through and lefttum lanes in order to obtain a strong correlation between observed and estimated safuration

T

It

flow rate. The discharge headway curve

is

presented in through lane under heterogeneous traffic to investigate saturation flow rate region.

Chapter

6 presents validation of the empirical model by using Synchro micro-

simulation software to calibnte the mcCified narameters of ohservcd data.

Chapter 7 discusses the policy implications of this study to the designers, planners

and researchers and their effects on planning, simulation, intersection design and signal design.

Chapter 8 provides

a conclusion and summary of the results obtained in

the

previous chapters of this dissertation. The implication of results is summarized in

this chapter. In addition, suggestions and recommendations for further needs are orovided.

research

I I I I t

.

Problem

.

stalements

. .

Objectives

Defending the criteria

of

the required case study

inlerseclions

Research

organization

'

Tme $hedu

le

of dala

T

I I I I I I

. Required software programs

. Define data accuacy l€vel

4,

Srtuntion Flow Reh Aorlysb

5.

l. U-turn adiustm€nt factor 2. Heavy vehicle adjustrnent fador

l I I t I I I I I

Inv6tigrtioo of Adjurtmmt Frcao]|

J. Number

oftlrough

lanes adjustment factor

4. Lane width adjustment factor

Conparison Study - compaxe results htween Makkah city and other cities - mmment of results

7. Conclusions,

Recornmendations for Futur€ Researches

Figure 1.2 Framework of the Research Study

I I I I t

Chapter

Literature Review

I I I

II

2.1 Introduction

Capacity of signalized intersections is the main factor affecting traffic performance

T

I I I I I I I I I I I I

and quality of service in urban roadway networks. Over the years, the factors that

affect intersection capacity have received considerable attention from traffic engineers and researchers around the world, especially during the last two decades.

Many factors had been investigated including driver characteristics, vehicle types,

control mechanisms of tum movements, traffic conditions and par:rmeters, and intersection geometry. Accordingly, this chapter presents a comprehensive review

on the previous studies in capaclty of signalized intersections, mainly saturation

flow rate and its adjustment factors which will be undertaken in this dissertation. The topics discussed in the chapter are as follows:

r

Saturation flow rate definition at signalized intersections and its relationship

with intersection capacity,

o

Base saturation

flow rate estimation methods.

r HCM 2000 methodology for estimating

saturation

adjustrnent factors, especially the factors that dissertation,

will

flow rate and its

be investigated in this

I I

I I I I I I t

Saturation flow rate modeling, and Previous studies in saturation flow rates at signalized intersections of Saudi Arabian cities.

2.2 Survey of SFR Related Intersection Capacity Saturation

flow rate is a basic parameter used to derive capacity of signalized

intersections.

It is determined

based upon minimum headway that the lane group

can sustain across the stop line. Saturation flow rate is computed for each of the lane groups. Saturation flow rate for the prevailing conditions can be determined

directly from the field measurements and be used without any adjusfinents.

If

default value is selected for the base saturation flow rate, it must be adjusted for

T

I t I I I I

various factors. This value reflects geometric, traIfic, and environmental conditions.

2.2.1 Definition of Saturation Flow Rate (SFR)

Saturation flow rate (SFR) is defined in HCM 2000 as "the equivalent hourly rate

which previously queued vehicles can traverse an intersection approach under prevailing conditions, assuming that the green signal is available at all times and no lost times are experienced". The unit for SFR is vehicle per hour per lane, which is measured during green phase.

The concept

T

I t I I

of

saturation flow as interpreted

in the Canadian Capacity

Guide

(CCG) is shown in Figure 2.1 (Teply 1995). The CCG defines saturation flow

as

"the rate at which the vehicles waiting in a queue during the red interval cross the stop line of a signalized intersection approach lane during the green interval". The

method

for calculating saturation flcw is

based on dcterrnining the headways

I I I I t I t t I I I I I I I

counting from the first to the last vehicle waiting in the queue, unlike the HCM 2000 which takes into account the headway of fourth to last vehicle in the waiting queue during red interval to determine the SFR. The CCG identifies a drop in saturation flow after about 30 seconds of green time. The guide also found the relationship between SFR values estimated by HCM and CCG to be as follows:

SH6ra

:

1.05 x 5966

(2.t)

Where, SHcv = SFR value estimated corresponding to HCM in pcphpl

Sccc

:

SFR calculated using Canadian Capacity Guide in pcphpl

SFR is mainly determined based on saturation headways of the departure queued vehicles downstream of the stop-line at a signalized approach of an intersection in

all traffic engineering guidelines. Deparhre headway is defined

as the

time interval

between two successive vehicles crossing a stopJine or any predetermined line at a

junction. The Saturation headway is defined as "the average headway passenger cars in a queue as they pass through a signalized intersection.

2500 2000

T

t I I I t

B 1500

1000 500

lr

o

0

05

10

15202530354045

Time from Start of Green (S)

Figure 2.1 Measured Saturation Flow Rate (Teply 1995)

10

between

I I I

I I I I I t I I I t I I I I t I t I

The concept and measurements of saturation flow were identifred as understood

and applied by different researchers all around the world. Differences among measurement methods suggested

in HCM

1985, CCG, and Aushalian Road

Research Board's (ARRB) special report on Traffrc Capacity and Timing Analysis

is

discussed by Teply and Jones (1991).

measured saturation

All

three studies suggest the use of

flow values rather than using the default values by diflerent

simulation progra:ms. The study defined saturation flow in accordance with HCM and CCG as listed before. ln addition, ARRB Report 123 defined saturation flow as

"the maximum constant depadure rates from the queue during the green time period, expressed in through-car unit per hour (tcu/hr)".

2.2.2 Relationship between Intersection Capacity and SFR

Generally, capacity at intersections is defined for each lane group. The lane group capacity is the maximum hourly rate at which vehicles can reasonably be estimated

to

pass through

the

intersections under prevailing traffic, roadway, and

signalization conditions

for specific lane type. Pammeters defining traffrc

conditions include volumes

on each

approach, distribution

of

vehicles by

movement (Ieft, through, right), vehicle type distribution, location and use of bus stops within intersection areas, pedestrian crossing flows, and parking movements

on approaches to the intersections. Roadway conditions include the basic geometry

of the intersections, including number and width of lanes, grade, aad lane

use

allocations. Signalization conditions include signal phasing, timing, and type of control. The cycle length is the total time (in seconds) required to complete one cycle ofphases. A phase is the part of the signal cycle allocated to any combination

of traffic movemeilts rcceivine ihe rieht-of-way sirru

l1

rlt41e",

'6ly

during one or more

I

l I I I I I I I I I I I I

I I I

intervals. Green time is defined as the duration of the green indication for a given movement at a signalized intersection. There is a strong relationship between the capacity of an approach at a signalized intersection and the measured saturation flow rate at this approach. The capacity a signalized intersection approach can be calculated using the

of

following equation:

.:gx(g/C)

(2.2)

Where, c = Capacity of approach in vehicle per hour (vph) S

:

Saturation flow in vehicle per hour of green (vphg)

g = Effective green time (seconds)

C

:

Cycle length (seconds)

The signal cycle for a given lane group has two simplified components: effective green and effective red. Effective green time is the time that may be used by

vehicles on the subject lane group at saturation flow rate. Effective red time is defined as the cycle length minus the effective green time. Equation 2.2 clearly shows that the capacity of intersection approach increases by increasing SFR. For clear understanding of this relationship, Figure 2.2 illustoates the vehicles' trajectory at an intersection approach. This figure is divided into three

parts. The first part shows a time-space plot

of vehicles on the

northbound

approach to the intersection. The intervals for the signal cycle are indicated in the

T

diagram. The second part repeats the timing intervals and labels the various time

I I t

intervals of interest with the symbols used throughout this chapter. The third part is

t2

t I I I I I t I t I I I I I t I I

I I

an idealized plot of flow rate passing the stop line, indicating how saturation flow is defined.

Figure 2,2 Attributes of Tralfic Flow at Signalized Intersection (HCM 2000)

The effective green time (g) is calculated as the actual green time (G) minus the total delay that occurs at a given approach of an intersection. The total delay time has two main parts; staxt-up delay time

(Tr) (the shadow area in Figure 2.2 Paft 3)

when green indicator turns on and the clearance delay time

which includes

delay time that occurs when yellow time staxts plus the all red time between successive phases. Thus, the effective green time can be calculated from the

following equation:

l t

(T)

IJ

I I t t I I I I I I I I I I t t I t I I

9:G-Tr-Tz

Where. g = Eflective green time, Tr

:

Start-up delay time, Tz

:

Clearance delay time

For more illustration for the haffic flow at sigrralized intersections, Figure 2.3 shows the flow rate of traffic during the green period from a saturated approach

(Stokes 1988). This figure shows a conceptual queue discharge model that is

typically used today. The area under the curve represents the total number of vehicles (i.e., total flow) discharged during the green and change intervals.

If

the

total flow is divided by the saturation flow rate (S), the resulting value is the effective green time (g). The difference between effective green time and sum of the green and yellow time is the total lost time. Several studies tackled the lost time

of signalized intersections. This topic is not covered here in literature review, since

it is out of the scope ofthis dissertation.

Lr i/

Effective Green Green time time

\i

Lz

Figure 2.3 Traflic Flow During the Green Period from a Saturated Approach

t4

T

(2.3)

I t I

2.2.3 Relationship between SFR and Headway

Headway

is

defined

in traffic

engineering handbooks as the time difference

T

between fronts of successive vehicle detection as shown in Figure 2.4. In the case

I I I I t I I

of signalized htersections, the departure headway is the most important parameter that affects directly the value ofSFR (S) and can be calculated as follows:

S

:3600 / h,

(2.4)

Where, S

= Saturation flow rate

h, = Average time of headway between the departure vehicles

during

saturation discharge condition.

i t I I I t I I I

Figure 2.4 Headway Determinations (Traffic Control Handbook 2005)

Figure 2.5 illustrates a queue of vehicles stopped at a signalized approach of an intersection. When the signal tums green, the queue begins to move. The headway

(or queue discharge interval), which is

successive vehicles, begins as they cross the stop

in

seconds between two

line (or curb line) of

the

intersection. The first headway would be the elapsed time, in seconds, between the start of the green light and the crossing of the rear of the first vehicle over the

reference linc. The second treadway would be the elapsed time between the

15

T

measured

t I I t I I t T

I I I

crossing of the rear of the first and second vehicles over the same line. Subsequent headways would be similarly measured.

o @ o---------Line of siehl for

headway

measurements

@

t L

h+t

3

h,+tl

'\

'2

Figure 2.5 Traffic Interruptions at Signalized Intersection Approach

The driver of the first vehicle in the queue must observe the signal change from red

to green and react to this change by taking hisArer foot off the brake, and accelerate

T

I I t t

through the intersection. Within the queue, the first headway

I I I t

be the

longest as a result of this process. The second vehicle in the queue follows a similar process, except that the reaction and acceleration period can partially occur while

the first vehicle begins to move. The second vehicle will be moving faster than the

first as it crosses the reference line, because it has an additional vehicle length in which to accelerate. Its headway will be generally less than that of the first vehicle. The third and fourth vehicles

T

will normally

will

each achieve a slightly lower headway than the

preceding vehicle. After a certain number of vehicles, "n", the effect of the start-up reaction and acceleration has dissipated, as shown in Figure 2.6 (Lee 1995).

16

I t t I I I

hr= cqlurqllon hcodwqy (scc) s = solurollon flow rtfe =t,600/h. (vPhSpl) lor.ilh.v.hicle (E6c) !tLt = Etort-up.delqy = .tortlng lost llna (sco) = tt* tz* tJ+ t4+ ts+ t3

5I7

T

I I I I t I

8 910 1lt2 13 1,1 15t8

Vehlclc ln oucuo

Figure 2.6 Saturated Headway Time and Starting Lost Time (Lee 1995)

After the effect of the start-up reaction and acceleration is removed, the successive vehicles move toward the stop line as a uniformly moving queue until the last vehicle in the original queue has passed. The headway for these vehicles will be relatively constant, which is denoted as "hr" (saturated headway time). Based on

HCM 2000, this saturated headway occurs after the fourth queued vehicle. Some researchers suggest that the saturation headway occurs after the

fifth or sixth

T

vehicle has stopped in the waiting queue. As shown in Figure 2.6, the effect of

I I I I

start- up and acceleration disappears after the sixth vehicle.

I I I

The determination methods

of

SFR were summarized

in terms of

headway

measurements in the HCM 2000, CCG, and ARRB as follows (I(hosla 2006):

r

HCM procedure describes measurements of saturation flow as starting from the fourth to the last vehicle in a waiting queue.

o

CCG procedure describes measurements of saturation flow as starting from the first to the last vehicle in a waiting queue.

17

I T

t I I I t I I I I t I I I t

o

ARI{B procedure (Report 123) divides each interval into three periods:

-

First ten seconds of green phase,

-

Remaining part of the green phase assumed to be saturated, and

-

Period after the end of the green phase, including yellow and red.

In addition, the PCU2 program developed by the University of Alberta was utilized

to

determine passenger car equivalent

t I

by using a least

estimated using CCG. Saturation flow values evaluated using HCM procedures were higher than those the PCU2 program but not sirnilar to those determined using

the ARRB procedure. The CCG reports a decline in saturation flow values after about 30 to 40 seconds ofthe green interval.

Over the past yeaxs, several studies tackled the relationship between the position vehicles in the queue and its departure headway time.

A

of

comparative study was

made between Yokohama in Japan and Dhaka in Bangladesh in terms of headway values ofthe passenger cars at signalized intersections (Rahman, Nur-ud-deen, and

Hassan 2005). Figure 2.7 illustrates the value

position in the queue.

It

of headway against the vehicle

shows the headway value decreases with increasing the

position number of the vehicle in the queue. In that study, the relationship between the observed headway and the location of queued vehicles from the stop line is also

in Figure 2.8. It illustrates that the headway time

decreases

with increasing the position number of the queued vehicle from the stop line position. Figures 2.7 and 2.8 show that the time headway values in Dhaka and Yokohama are significantly different. Headway observed in Dhaka is lower than

T 18

T

squares optimization

technique. The ARRB method found saturation flow values much higher than those

presented as shown

T

separate

that of Yokohamq even under the same taffic conditions which highlight the imporunce of studying SFR under the local conditions ofeach city.

3.5

!t? a; 7

E

z.s

& E

9.2 4 5 6 7 8 910 llt2 Position of Vehicles in Queue X'igure 2.7 Comparison of Headway Values in Dhaka and Yokohama

45

F40 €35

Fe, Bg2s l, !.

!d

.E

20

rs

g5

ro

Fl

0

I

I I I

56789

l0 1l 12

Queue Position

X'igure 2.8 Location of Queued Vehicle in Dhaka & Yokohana

t9

13

I T

t I I I I I I I I I I I I t I I t I I

The same fact has been observed in many other studies of different cities all over the world, such as that by Gerlough and Wagner (1967), Lee and Chen (1986), Al-

Ghamdi (1999), and Lee and Do (2002). The results of these studies in terms of the relationship between the position number of the vehicle and its departure headway are shown in Figure 2.9.

All of

these studies prove that the mean value of the

departure headway decreases with increasing the position number of the queue vehicle. However, this relationship is not similar over these studies. Recently, Jin et al. (2009) conducted elaborate study on headway distribution in

Beijing. Here, the detailed distribution of the departure headways was carefully examined by using video traffic data collected from Beijing between 2006 and

2007. The results show that the departure headways for each position follow a certain log-normal distribution. The statistical results of the observed headways are

summarized in Table 2.1, which shows that both mean and variance of observed headways decrease with increasing a vehicle's position number in the queue.

E

4

o

3.5

qrc

>

.:>

3P

& waenert -Gerloueh . Lee & ehen (19tr6)

*"

3

1967)

Al-Ghamdi ( 1999, two-lane) Al-Ghamdi ( 1999, three-lane) Lee & Do (2002)

2.5 2

\a

l.) I

x

0.5 0

| 2 3 4 5 6 7 8 910 lt t2t3t4

15 16

l7

18 1920

Queue Position

Figure 2.9 Departure Headways at Signalized Intersection in Dilferent Studies TABLE 2.1 Departure Headways (seconds) as observed in Beijing

20

I I I I I

Position Sample size Maximum Minimum Range Mean Median Variance 1

+zJ

9.74

1.26

8.48

4.53

4.26

1.51

2

.+zJ

8.36

1.05

7.31

3.05

2.81

1.08

3

+L)

6.69

1.05

5.64

2.76

2.56

0.93

4

^.t.\

9.32

1.05

8.27

2.57

2.39

0.90

5

416

7.77

1.05

6.72

2.51

2.35

0.88

6

394

K)A

1.05

5.21

2.36

2.23

0.78

7

298

4.91

1.05

3.86

2.19

2.r0

0.69

8

t46

5.25

1.05

4.16

2.04

1.89

0.66

a

42

3.15

1.05

2.10

1.87

1.83

0.56

T

I I t I t I

l t

2.3 Review of the Base Saturation Flow Value Estimation Use of inaccurate departure headways may lead to wrong assessment of delays and

thus results in inefficient design of traffic signals. Due to their importance, many investigations on saturation flow rate have been carried out in the last few decades. These studies investigated the statistics

of departure

headways conceming some

external influence factors. The base value of SFR is usually used as an ideal value

T

I I I I t I

when no extemal factors affect the departure time headways of vehicles. Ideal conditions in the HCM 2000 assume the followine:

o

3.6 meter lane width,

r

No heavy vehicles,

r

Flat gradient,

.

No parking or bus stops near the intersection, no pedestrians or cyclists,

21

I t

I I

Uniform movement type, i.e. only shaight or turning movement,

I I I I I I I I I I I I I I

traffrc lanes and exclusive right-tum lane in the other approach) in

T

I I

Exclusive left-tum lane and protected left-tum phasing, and

The selected street should have at least three trafhc lanes (including through each

direction.

The HCM 2000 prescribes an ideal (base) saturation flow rate

of

1900 pcphpl

which corresponds to saturation headway of 1.9 sec. However, the previous version

ofHCM

1985 states the saturation flow rate as 1800 pcphpl, which determines the

fluctuation behavior of this value, even in the same country. Furthermore, the Pennsylvania Departrnent saturation flow rate

of

of Transportation (PENNDOT) currently uses a

base

of

1900

1800 pcphgpl, which is less than the default value

pcphgpl provided by HCM 2000. This is

to

account

for the less aggressive

characteristics of the local drivers.

In professional literature, there are many studies that address the estimation value of the base saturation flow at signalized intersections. The results of some of these studies are presented here. The saturation flow of pervious studies was summarized as shown

A

in Table 2.2 (Tumer and Hatahap 1993).

comprehensive study was conducted

by

Texas Transportation Institute

(Bonneson et al. 2005). In this study, the authors investigated the effect of heavy vehicles, speed limit, and traffic pressure. They found that the saturation flow mte under the ideal conditions

is 1905 pclh/ln that it

decreases

by 9 pclh/ln for each

single mph decrease in the speed limit. An approach with two through lanes has a saturation flow rate of 130 pc/h/ln higher than that ofan approach with one through lane.

22

I

I t

TABLE 2.2 Previous Studies' Saturation Flow Rates Study

Country

Mean (pcll/ln)

Webster & Cobbe

UK

1800

Kimber et al.

UK

2080

Branston

UK

1778

Miller

Australia

1710

Indonesia HCM (Binkot)

Indonesia

2t96

H.E.L.Athens

Greece

r972

Shoukry & Huizayyrn

Egvpt

t6t'7

Hussain

Malaysia

1945

Coevman & Meelv

Chile

1603

India

1232

Arahan Teknik

Malaysia

1904

De Andrade

Brazil

1660

T

I I I t I I I t I I I I I

Bhattacharva

& Bhattacharva

The saturation flow rate was investigated over several intersections in the city of Stellenbosch

in South Africa (Bester and Meyers 2007). The results of the

study

are summarized in Table 2.3. This study shows that saturation flow rate in South

Africa is much higher than in other counties. This could be an indication of the aggressiveness

oflocal drivers. It also shows that an increase in the speed limit

and

number of though lanes leads an increase in saturation flow rates. In addition, an

T

increase in the gradient leads to a decrease in the saturation flow rate. Finally, the

I I I

effects of speed limit, gradient, and number of through lanes on the saturation flow rute arc much higher locally than in USA.

23

I

I I I t I

I I t I I I I t I ! I I

t I I

TABLE 2.3 Saturation Flow Rates (vphpl) in Stellenbosch, South Africa Sample Saturation Standard

Minimum Maximum

Intersection Size

flow mte deviation

Dorp/Strand

214

2026

133

1839

2254

Moltn/Bird

74

1711

I JJ

1565

1946

Adam Tas/Bird

103

1820

151

1625

2071

Strand/Van Reed (-G)

ll8

2197

204

1879

2471

Strand/Van Reed (+G)

140

2044

t23

1908

2314

PaarVWelsevonden

84

2000

140

1835

2342

Webersvallei/R44 GG)

102

2370

148

2062

2605

Webersvallei/R44 (-G)

105

2076

267

1553

2516

Strand/Van Reede (Right)

99

1840

180

1814

2069

Strand/Saffraan (Right)

95

1920

190

1603

2195

2.4 HCM 2000 Method for Estimating SFR and its Adjustment Factors

HCM 2000 is considered as one of the most widely used traffic engineering guidelines around the world. HCM 2000 recommends measuring saturation flow rate on a lane-byJane basis through observations of headways as vehicles pass over

the stop line of the intersection approach. According to the HCM methodology for sienalized intersections, satuation flow is defined as follows:

S

:

SoxNxf*xf"vx

fgx

fpxfruxfuxfnrxfr-r

Where, S=

prevailing saturation flow rate in the lane group. rphg;

24

(2.s)

I

I I t I I I I I I t I I t T

T

I t I I I

:

So

ideal satuation flow rate per lane, (:1900 pcphgpl);

N = number of lanes in the lane group;

:

f,,

adjustment factor for lane width;

:

fsy

adjustment factor for heavy vehicles;

f,

:

adjustrnent factor for grade;

fo

:

adjustment factor for parking;

f65

fu

= adjustment factor for local bus blockage;

:

adjustment factor for area type;

fp1 = adjustrnent factor for right-turns;

flr

= adjustment factor for left-turns.

Since one of the major tasks of this disse(ation is to investigate some of these

factors in Makkah, an intensive review on the previous studies that tackled this

point is presented in the following subsections, especially the factors that will be taken into consideration in this dissertation.

2.4.1 Lane Width Adjustment Factor,

f*

This factor depends on the average width of traffic lanes in a group. It is used to account for both the reduction in saturation flow rates when lane widths are less

than 12 ft (3.65 m) and the increase in saturation flow rates when lane widths are greater than 12 ft. The adjustment factors of lane width as presented in HCM 2000 are shown in Table 2.4. Lane width factors should not be computed for lanes that

are less than 8

ft. wide. A lane width of 12 ft (3.65 m) would result in

an

adjustrnent factor of one, which would have no effect on saturation flow rate. A lane width of less than 12

ft results in a factor

less than one, thus lowering the

I T

I I I I t I I I I I I

satwation flow rate, and a lane width greater than 12 ft results in a factor that is greater than one, thus increasing the saturation

HCM aside, many researchers attempted to find the impact of the lane width on SFR value.

A

study by Agent and Crabtree (1983) indicates that lane width does

not have an effect on saturation flow for lane widths of 3.0 m (10 ft) or more. For lane widths between 2.7 and3.0 m (9 and

T

I t t I

l0 ft),

a 5 percent reduction in saturation

flow was found as compared to lane widths of 3.0 m (10 ft) or more. No lane widths below 2.7

m (9 ft) were surveyed in the

study. There was

a

slight

unexplained reduction in saturation flow for lane widths greater than 4.5 m (15 ft).

A similar analysis was

performed with the limited data available for commercial

vehicles, and no effect was found even for lane widths below 3.0 m (10 ft). Table

2.5 illustrates the effect of lane width on saturation flow found bv Asent

and

Crabtree.

TABLE 2.4 Lane Width Adjustment Factor of SFR Based on HCM 2000 Average Lane Width, w (ft)

Lane Width Factor, f,"

8

0.867

o

0.900

10

0.933

ll

0.967

t2

1.000

13

1.033

t4

1.067

15

1.100

T

t I

flow rate.

zo

I

t I I I

TABLE 2.5 Effect of Lane Width on SFR (Agent and Crabtree 1983) Lane width

Total Headway

Average headway

Saturation flow

(ft)

(sec)

(sec)

(vphgpl)

9- 9.9

8s8

2.29

r572

-

10.9

2839

2.16

1667

rt-

12.9

I 1089

2.18

1651

-

14.9

2454

2.18

l65l

more

690

2.21

1629

14.9

16382

2.18

1654

17062

2.18

1653

10

I t I

13

I 5 or

10

I t I I t t I

I I t

I I t

-

l0 or more

In the study by Zegeer (1986), saturation flow rates were evaluated on approaches

with lane widths varying between 2.6 and 4.7 m (8.5 and 15.5 ft). Saturation flow data were collected from 2,733 vehicles on eleven approaches with lane widths

varying between 2.6 and 2.9 m (8.5 and 9.5 ft). Four approaches with lane widths

varying between 3.9 and 4.7 m (13.0 and 15.5 ft) were also surveyed, with sample size

a

of 1,568 saturation flow vehicles. All baseline conditions except for

lane width were held constant at these locations. The survey results were then compared with those of the baseline condition surveys (with a sample size of 6,687

saturation flow vehicles). The narrower lane widths demonstrated saturation flow rates between 2 and 5 percent less than did those in the baseline suweys, while the

wider lane widths demonstrated saturation flow rates 5 percent greater than did those

in the baseline surveys. Table 2.6

adjustment factors (Zegeer 1986).

27

summarizes the proposed lane width

I T

I

TABLE 2.6 Effect of Lane Width on SFR (Zegeer 1986) Lane width cateCory (ft)

Saturation flow adiustment factor

8-8.9

0.95

-9.9

0.98

10

-

12.9

1.00

13

- 15.9

1.05

T T

I t I I t I

9

The relationship between lane width and saturation flow rate was investigated on urban and suburban signalized intersection approaches using fleld study sites with the most ideal conditions possible (Potts et al. 2007). The research indicates that

the satwation flow rate varies with lane width.

A

total of 25 intersection

approaches (exclusive through lanes only) located in nine cities within five states

T

(Florid4 Arizona, Oregon, Idaho, and Washington D.C) were included in the study.

I

There were 1,199 average headway samples (same as "cycles" or "observations") recorded at all sites. These were distributed by queued vehicle positions as follows:

T

o

8d queue position: 364 measurements (30.4 percent)

I I I

r

9ft queue position: 110 measurements (9.2 percent)

.

10'n queue

.

1Ifr queue position: 3 measurements (0.25 percent)

t t t I

position:722 measurements (60.2 percent)

Due to the extremely smail number of vehicles at the 11'n queue position, those

three measurements were excluded from subsequent analyses. The study sites consisted of six lane widths groups; 2.9 m (9.5 ft) (9 study sites); 3.3 m (11 ft) (5

28

t T

I I t I I I I I t

i t I t I I t I I I

study sites), 3.5 m (1 1.5 ft) (1 study site), 3.6 m (12 ft) (6 study sites), 4.0 m (13 ft)

(3 study sites), and 4.3 m (14 ft) (1 study site). The average satwation flow rate was found to be in the range from 1,736 to 1,752 pc/h./ln for 2.9-m (9.5-ft) lanes, in the range from 1,815 to 1,830 pclh./ln for 3.3- to

3.6-m (11-

to 12-ft) lanes, and in the range from

1,898

to 1,913 pc/h/l for

lane

widths of 4.0 m (13 ft) or greater. These measured saturation flow rate values are generally lower than those cunently used in the HCM. Furthemtore, the percentage difference in the saturation flow rate between sites with 2.9 and 3.6 m (9.5 and 12

ft) lanes was found to be about half the value used in the HCM. Since the data were limited to queue lengths between 8 and 11 vehicles, the research results do not directly address queue lengths greater than lane widths 2.9 13 ft, and

vehicles. The adjustment factor

m,3.3 m, 3.5 m,3.6 m,4.0 m, and 4.3 m (9.5 ft,

l4 ft) are 0.91,0.96,0.96,0.96,

The results

ll

11

ft,

11.5

of

ft,12ft,

1.0, and 1.0 respectively.

ofa saturation flow study by Lewis and Benekohal (2007) presented for

25 through lanes at signalized intersections in Panama City, Republic of Panama.

Only intersections with variations in lane width and percentage of heary vehicles were considered. The rest of the factors which can aflect the driver's behavior were considered to be under ideal conditions.

All

studied sites are one-through lane only.

Lane widths of2.80, 3.00, 3.35, 3.65 and 4.00 m were considered. Percentages

of

heavy vehicles of 0, 10, 20,30 and 40Vo were considered. A video camera was used

to capture the traffrc data, which was collected from I I am to 3 pm on weekdays. The total period

of 100 hours was included in the study analysis. A sample of data

of 200 cycles was validated for all sites (40 cycles per site "intersection"). The adjustment factors of lane widths 2.8 m, 3.00 m, 3.35 m, 3.65 m, and 4.00 m were 0.82, 0.93, 0.98, 1.t1, and

l.i i

respectively.

29

I I

2.4.2 He*^y Vehicle Adjustment Factor, fsy

T

in the specified lane group. This factor

I I I t t I

reduction in saturation flow rate due to the presence of healy vehicles in the traffic

The heavy vehicle adjustment factor is related to the percentage of heavy vehicles accounts

for the additional delay and

stream. In this study, a heavy vehicle is defined as any vehicle that has more than

four wheels touching the pavement. The additional delay and reduction in saturation

flow are due mainly to the

difference between the operational

capabilities ofthe heavy vehicles and passenger cars and the additional space taken

up by heary vehicles. Consequently, the presence of these vehicles in the traffic stream increases the headways and lost time, making less effective use of green

time and adding to congestion and delays. In addition, the presence of a heary truck in front of a passenger car causes the passenger-car driver to be more cautious

ofthe healy truck's large size and the resulting diminished sight

T

because

t t

This causes the headways of passenger cars to be larger and this increase must

T

distances.

therefore be considered in the overall capacity reduction due to healy trucks. In addition, the adjustment factor for hear,y vehicles is used to account for the effects

of no-turn heavy vehicles to no-tum passenger cars in through lane group. HCM 2000 proposed adjustment factors of heavy vehicle in the saturation flow rate as

I I I

between successive vehicles

T

using a video camera. The sampling periods were during moming and evening

I

peak hours, when the intersections were saturated with regular traffic and heavy

shown in Table 2.7.

The effects of healy vehicles in through lanes and

investigated at two intersections (4Jeg) in Taipei (Tsao and Chu 1995). Headways

in through and in left+um lanes were recorded by

vehicies. A szrnple of2042 hean

{>

./x

I'J

Observed SFR (pcphpl) [o]

1708

1780

1958

2066

2016

Estimated SFR (pcphpl) [e]

1770

1680

2002

1849

1930

HCM 2000 SFR (pcphpl)

1645

1645

1493

1508

1715

e]

-fJz

100

-44

217

86

Deviation' [dile]

2.r7

5.95

0.97

25.46

3.83

Deviation [o

-

Chi-Square Test

t

38.38

=Lla2hl,ar=

q

1t7

T

is 1.54 sec,2.07

l)

I I

I I I I I I I I I I I I

I t I

I I I I

The calculated value of (1' = 38.4) in Table 5.6 is higher than the critical values at

all levels of significance. It implies that the observed SFR is significantly different

from estimated SFR in through lanes as calculated using Model I . The null hlpothesis, which states that there

is no significant

estimated and observed results (using Model

difference between the

l), is rejected.

The mean value of headway categories does not replicate the best model of the headway pattem under the heterogeneous traffic. Then,

headway distribution

is required to examine performance of various headway

models, investigate a better goodness

for

a certain

a detailed analysis of

offit

tests, and select the most suitable model

traffic conditions or situations.

It is very difficult to simulate the discharge headway of individual queued vehicle because

of the great variations in the driver behaviors, vehicle characteristics

and

traffrc environment. The distribution ofvehicle headway is Iirndamental for several important traffic flow theory research and simulation issues. Accurate modeling and analysis of vehicle headway dishibution helps traffic engineers to maximize

roadway capacity and minimize vehicle delays. Additionally, vehicle headway

to

distribution

is

movements.

It is essential for determining

intersections as

closely related

vehicle merging and permitted left-tum

the capacity of roundabouts

and

well as adjusting or coordinating signal timing plans at

the

signalized intersections. Furthermore, proper headway models can be applied to generate vehicle events

in microscopic simulation models and to traffic

safety

analysis. A probabilistic approach based on first-order second-moment method has

been adopted

to estimate the saturation flow and the delay

caused to traffic, at

signalized intersections, under heterogeneous traffic conditions (Arasan and Jaggedness 1995). The studv b1- Arasan and Kosh;r (2003) found that thc negative

ll8

I I exponential distribution is adequate to model headways over the entire range of

T

traffic flow considered. The study by Gorualez (2006) found that the average

I

headway for the city

i I

of Monterrey, Mexico is slightly higher to that of

conditions. Two or three vehicles incur start-up lost time. instead

US

offive in United

States.

Hwang et

d.

(2005) suggested that the discrepancy between observed and HCM

of

I

2000 approaches might be caused by the use

T

information to capacity analysis they suggested use of disaggregated headway data

I I t

for better understanding of relationship between time headway and speed. It is

I t I I

distribution, which

i I

I

assessment

inappropriate headway in the

of adjustment factors. Emphasizing on the use of precise headway

suspected that the mean headway, used for estimation of the adjustment factors,

might be biased by the few large values.

It is important to identifr an appropriate representative is

value of the headway

sometimes different from the mean value, because it

describes a realistic headway pattem

of heterogeneous traffic. The

appropriate

headway value for fitting the distribution in this study was selected by observing

the smoothness of the shape of the histogram. Hence, the statistical software package Minitab 12 for Windows, was used to plot the frequency histograms of all the data sets by displaying the descriptive statistics.

The histogram of the headway distribution for the different headway categories shows the percentage of the vehicles passing at a particular headway. Figures 5.5

through 5.8 show the histogram categories Opp, hpH,

h11p,

of the headway distribution for the headway

and hgs) in the through lanes respectively.

The maximum percentage ofvehicles for each headway category (hpp, hpn, hnp, and

T

h,JH)

is ohtained from thc follcr.r'ing figures. ThsS'

I

I

ll9

elre

7

.29/c, 11.4V', 16.3Vo,

md

I t 12.50%

I I

I

I I t T

I t I I t I I I I ! t t

respectively. Theses percentages of vehicles move at the headway

of

1.40

sec, 1.71 sec, 1.97 sec, and 2.02 seconds respectively. These headway values correspond to the mode values of sample set fbr each headway category (hpp, hpn, hnp, and hsu).

8

7

6

o, E+ 6" 2 1

0

c.l o o, 0o ra)F-ooat+\ooooc!st o -j-j o ct -

\o

ra| -:

t .a Jc.i

a.l

oi

oi

Headway (sec)

Figure 5.5 Histogram of Headway Category lpp in Through Lanes

4.. 00

.a a!

O\lr) (.)0\

.-.t

-i

qq-q al ol

.o

cn

\ocl€ c.rt\f

\.n

Headway (sec)

Figure 5,6 Histogram of Headway Category lrps in Through Lanes

t20

l8

l6 t4 l')

^

Ito f8 {)

4 2

0

YS