Click for next page ( 32


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 31
31 Chapler Five SATURATION HEADWAYS IMPORTANCE OF SATURATION HEADWAYS The basic parameter used to compute intersection capacity is saturation headway. For signalized intersections, ideal saturation headway is the difference in the passage time at the intersection stop line between two consecutive vehicles once the queue is moving in a stable manner. The 1985 Highway Capacity Manual (HCM) notes that the saturation headway is "estimated as the constant average headway between vehicles which occurs after the fourth vehicle In the queue and continues until the last vehicle In the queue clears Hat intersection." Field measurements must consider the start up lost time, or that time at He beginning of He green phase Hat is required for the queue to begin to move. The capacity procedures given Chapter Nine of the HCM (Signalized Intersections) provide a standard value for the ideal saturation headway of 1.9 seconds per vehicle, which yields an ideal saturation flow rate of 1900 vehicles per hour of green. The procedure provides adjustments to this ideal value to consider the effects of intersection geometry, opposing traffic Bow, signal liming parameters, and pedestrian Bows. For TWSC intersections, Me folZow~up fume, or lbe time headway between the second and following vehicles entenngirom a continuous queue Ante minor stream and utilizing He same major stream gap, is equivalent to the saturation headway. FoDow-up time values published In the 1994 update to the HCM range from 2.1 seconds for major street left turn vehicles to 3.4 seconds for minor street left turn vehicles. Recommended values from this current study range from 2.2 seconds to 3.5 seconds for these same vehicle movements. For AWSC intersections, the saturation headway is the time headway between two vehicles departing from the same lane under conditions of continuous queueig. Saturation headway is used to compute capacity In both He 1985 HCM and the 1994 HCM update. Saturation headway is also one of the primary input parameters marred for He proposed capacity procedure descnbed in chapter three of this report RESEARC:EI HYPOTHESES: FACTORS AFFECTING SATURATION HEADWAYS Past studies and close observation of traffic operations as part of this study provide evidence that the saturation headway for a vehicle at an AWSC intersection is a function of both traffic flow characteristics as weD as intersection geometry. Since the saturation headway is the key parameter to be used in the estimation of He approach capacity, it is important to understand and quantify the factors that affect it. Based on previous work, eight degree of conflict cases are assumed to be necessary to completely describe He conditions faced by a driver at the stop line of an AWSC intersection. Four hypotheses are proposed and evaluated. Hypothesis ~ The saturation headway of a subject vehicle is dependent upon the degree of conflict faced by He subject driver as measured by the presence of vehicles on the opposing and conflicting approaches. la. The saturation headway for case 2 is larger Can for case I. Ib. The saturation headways for case 3 and case 4 are the same and bode are larger than He values for case 2. Ic. The saturation headways for cases 5, 6, and 7 are the same and are all larger than He values for cases 3 and 4. Id. The saturation headway for case ~ is larger An the values for cases 5, 6, and 7. Hypothesis 2 The saturation headway of the subject vehicle is dependent on intersection geometry, particularly the number of lanes on the conflicting approaches, He opposing approach, and the subject approach. 2a. The saturation headways for Group ~ sites are less than He values for Group 2 sites. The saturation headways for Groups 3 and 4 sites are not significanfdy different, are greater then the values for Groups ~ and 2, and are less than the values for Groups 5 and 6.

OCR for page 31
32 2c. The saturation headways for Group 5 are greater than the values for Groups ~ Trough 4 and are less Man for Group 6. 20. The saturation headway for Group 6 are greater than the values for Groups 2 through 5. Hypothesis 3 The saturation headway of the subject vehicle is dependent on in directional movement as wed as the directional movement of the opposing and conflicting vehicles. 3a. The saturation headway for left turn vehicles is greater than He value for Trough vehicles. 3b. The saturation headway for left turn vehicles is greater than He value for right turn vehicles. 3c. The saturation headway for Trough vehicles is greater than the value for right turn vehicles. 34. The specific interactive combinations of the subject, opposing, and conflicting vehicles affect He saturation headway. Or, the greater He conflict between movements, the higher the saturation headway is likely to be. Hypothesis 4 The saturation headway of He subject vehicle is dependent upon its vehicle type. 4a. The samradon headway for passenger cars is less than for heavy vehicles. These hypotheses are tested In two ways. First, difference of means tests are conducted using the sample means, sample variances, and sample sizes from the samples being compared. The z-statistic is evaluated at the 0.01 significance level In each case. Second, regression equations are developed to help assess Be quantitative effects of these factors. HYPOTHESIS TESTIN~GROIJP 1 SITES Analysis of Eight Degree of Conflict Cases It was ~n~dady proposed Hat eight degree of conflict cases are required to Filly descnbe the conditions faced by the subject approach driver. These cases are described Table 35. Table 35. Headway Cases Table 36 shows the mean headway, standard deviation, and number of observations for each of the eight degree- of Conflict cases for the single lane (Group I) sites. The mean and plus/minus one standard deviation range are Gown In Figure 9. Table 36. Saturation Headways, Single Lane Approach Sites 3.86 4.78 S.8g S.89 7.42 7.34 7.3S 9.SO 1.38 t.S6 i.70 i.67 1.95 2.~! 2.09 2.92 385S te74 1073 934 s72 572 S55 427 The data presented Table 36 and in Figure 9 suggest that separate headway cases should be maintained for case I, case 2, and case S. However, He data suggest Cat cases 3 and 4 and cases 5, 6 and 7 might be combined. Difference of means tests were completed to determine statistically if the hypotheses la through le can be supported. Table 37 shows the results of these tests.

OCR for page 31
33 Table 37. Hypothesis Tests: Consolidation of Headway Cases Stan Headway Case Mean Vanance Observations . Z Test Results Computed z value PI =z) one-tail z Cntical one-tail P(T < =z) tw~tai1 z Cntical tw~tail ID _ 3.86 1.90 3855 2 4.78 2.43 1874 -21.68 0.00 2.33 0.00 2.58 Reiect 3 5.88 2.89 1073 4 5.89 2.79 934 4.15 0.44 2.33 0.22 2.58 Accent 2 4.78 2.43 1874 5.88 2.89 1073 -17.39 0.00 2.33 0.00 2.58 Re ect J Table 38. Hypothesis Tests: Consolidation of Headway Cases Stan Headway Case Mean Vanance Observations Z Test Results Computed z value P(Z<=z) one-tail z Cntical one-tail P(Z< =Z) tw~tail z Critical tw~tail Ho: Cases ale identi=1 3 4 5 6 5 5.88 5.89 7.42 7.34 7.42 2.89 2.79 3.80 4.45 3.80 1073 934 572 572 572 .15 0.44 2.33 0.22 2.58 Accept 0.65 0.26 2.33 0.13 2.58 Accept 7 7.35 4.37 555 6 7.34 4.45 572 7.35 4.37 555 7.35 4.37 555 8 9.50 8.53 427 ' 1 ' 1 ' 1 0.59 0.28 2.33 0.14 2.58 Accept -0.06 0.48 2.33 0.24 2.58 Accept -12.91 0.00 2.33 0.00 2.58 Reiect AWSC- Single Lane Sifes 15 ~0 5 o i: ........... ........... = = 1 2 3 4 5 Case 6 7 8 Figure 9. Mean Values and Plus/Minus One Standard Deviation Saturation Headway Values Three major conclusions can be drawn from Table 38: . . There is statistical evidence that headway cases 3 and4 can tee combinedinto one case. This means that the direction of approach of a vehicle on the conflicting approach does not significantly affect the saturation headway of the subject approach vehicle. There is statistical evidence that headway cases 5, 6, and 7 can be combined into one case. This means that when faced with two vehicles on Me opposing and conflicting approaches, the direction of approach is not significant. Cases I, 2, and g should remain as separate cases. The combined cases are shown in Table 39. Note: From

OCR for page 31
34 this point on in this report, the degree of conflict cases wiR be designated from ~ to 5, according to the results presented In this section of the report. Table 39. Saturation Headway Dam for each Case 3.86 4.78 S.88 7.37 g so 1~8 1.S6 1.69 2.0S 2 92 38SS 1874 2007 1699 427 Note: StDev is He standard deviation Obs is the number of observations. Figure 10 shows Me frequency distribution for the saturation headway measurements for each degree of conflict case. Figure 10 shows that the more complex the degree of conflict,` the higher the mean value of saturation headway and the hither the vanability about Me mean, bow important conclusions. Note also that Me headway dis ribudons are not normal, but have a definite skew to Me right. The variation of Me mean value of Me saburabon headway at each site for each of Me new degree~f- conBict cases is shown in Figures I} Trough 15. These figures illustrate a high level of agreement (relatively low vanadon) between the measurM values at each site. Also shown in Me figures are the confuted capacities (3600 divided by Me samradon headway3 for each site. 0.10 ~ . =0.05 0.00 . ~-,`. 0 5 f0 15 20 Satumffon Headway | |-N1--N2- N3--N4--NS| l Figure 10. Frequency Distributions for Degree of Conflict Cases 15 ^10 S O Saturation Headways, Case Single Lane Sites Site ~ H - dwarfs Capacity 1200 800 400 . -o Figure 11. Variation of Saturation Headway and Capacity by Site, Saturation Headways, Case 2 Sing/e Lane Sites 15 ,4 10 :c 5- O - ~,EN,~,~I,t~,~3,~,tNI~,~,~I~l O Site Headway Capacity - 1200 ........................................................................... ........................................................................... 800 400 Figure 12. Variation of Satumbon Headway and Capacity by Site, 15 ~10 s 5 O Saturation Headways, Case 3 Sing/e Lane Sites ~3 Headword Capacity 1200 800 400 O Figure 13. Variation of Saturation Headway and Capacity by Site, Case 3

OCR for page 31
15 ,... ~ . . + l ~ Site Headway Capacity 120lv 800 -400 -O | ~ Headword Capacity Are 14. Variation of Saturation Headway and Capacity e4 Saturation Heac~ways, Case tSingle Lane Sites 15 , . ~0 ~ ~.~.~.~..~.~.~.~....... . .... ~ . . . I.' .... id, .... . _ : . . . - -I--- ~1- ~ - T _ ~ ~ ~ ~ ~ Site Headway Capacity 1200 800 400 O ~ 15. Variation of Saturation Headway and Capacity by Site, Case 5 .. ... Table 41. Headway Case 1-Turning Movement Difference in Means Tests Effect of Directional Movement. Table 40 Case 1 Mean Variance Ob~ons 3.92 2.S6 367 3.92 1.74 3063 3.35 2.31 425 Case 2 Mean Variance Ob~ons S.S9 2.69 197 4.72 2.28 14S6 4.45 2.62 221 Case 3 Mean Variance Ob~ons 6.02 2.92 341 S.97 2.66 1373 S.32 3.31 293 Case 4 Mean Variance Observations 8.00 S.11 32S 7.34 3.84 1148 6.63 3.65 226 Case 5 Me" Variance Observations 9.94 7.84 106 9.31 6.45 270 9.58 20.70 S1 The difference in means tests for directional movement of Me subject vehicle are summarized in Tables 41 through 45. .^ , .. 4 ..~. ... 1' t:.:-::--:::::-:.:-:-:-:::.:::::-:::::.::-::::::::::::::::::::::::: :::::::::::: :::::-::: :::-::::::-:-:-:-: ::-:: E:::::::::::::::::: -:::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::- :::::::::::::::::: :::::::::::: ::::::::::::::: -:::::.:.- :.:.:::.:::.:::.:. -::::::::::::::::::::::::::: Ista~cs I I T T I 1~ T=g Movement LT ~111 Me" 3.92 1 3.92 Variance 2.S6 1 L74 Observations 367 1 3063 Z Test Results Computed z value -0.04 P(Z<=z) one-tail 0.49 z Critical one-tail 2.33 P(Z<=z) ~v~tai1 0.24 z Craft ~~1 2.Sg Ho: Cases are identical Accent LT 3.92 2.S6 367 S.1S 0.00 1 2.33 1 Coo 1 2.S8 Reject | RT 3.3S 2.31 42S TH 3.92 1.74 3063 7.44 0.00 2.33 0.00 2.Sg | Reject RT 3.3S 2.31 42S

OCR for page 31
36 Table 42. Headway Case 2-Turning Movement Difference in Means Tests alel~ ler 1 ' ''""""'' '~"' "" ''''""'''''' 'I' '' '''' '' ' ' ' ' ''' ''''' '' ''''''] - - s~C5 T T T I T I L Turning Movement LT TH LT RT TH Mean S.S9 4.72 S.S9 4.4S ~4.72 Vanance 2.69 2.28 2.69 2.62 2.28 Observations 19 7 1 4S 6 19 7 2 2 1 14 S 6 Z Test Results Computed z value 7.08 7.11 P(Z<=z) one-tail 0.00 0.00 z Critical one-tail 2.33 2.33 P(Z<=z) two-tail 0.00 0.00 z Critical two-tail 2.S8 2.S8 H0: Cases are identical Reject ReJect RT 4.45 2.62 221 2.27 0.01 2.33 0.01 2.Sg tccept Table 43. Headway Case 3-Turning Movement Difference in Means Tests ... , , ........................................................... Staffsffcs Tuming Movement LT TH LT RT TH Mean 6.02 S.97 6.02 S.32 S.97 Variance 2.92 2.66 2.92 3.31 2.66 Observations 341 1373 341 293 1373 Z Test Results Computed z value 0.S6 4.96 P(Z<=z) one-tail 0.29 0.00 z Cntical one-tail 2.33 2.33 P(Z<=z) two-tail 0.14 0.00 z Critical two-tail 2.SS 2.S8 H0: Cases are identical Accept R~ect ~- Table 44. Headway Case 4-Turning Movement Difference in Means Tests TH 7.34 3.84 1148 RT S.32 3.31 293 S.57 0.00 2.33 0.00 2.Sg Reject P :::: '''-'' '''''''' ''' :$ ~ Stadstics 1 1 1 1 1 1 Tuming Movement LT TH LT RT ~ Mean 8.00 7.34 8.00 6.63 1 Variance S.11 3.84 S.11 3.6S Obsenrations 32S 1148 32S 226 Z Test Results Computed z value 4.83 7.68 ~ P(Z<=z) one-tail 0.00 0.00 | z Cntical one-tail 2.33 2.33 p(z<=z)two tall 0.00 0.00 zC:riticaltwo-tail 2.S8 2.S8 H0: Cases are identical Reject R~ect S.04 0.00 2.33 0.00 2.S8 Reject RT 6.63 3.65 226

OCR for page 31
37 Table 45. Headway Case 5-Tuning Movement Difference in Means Tests 3 .................................................................................... '"''' ''""''''''"'"''""'"'' ''''""" :' ''' , ~. ~, , ; Staff~ffcs Tu~ning Movement LT TH LT RT TH RT Mean 9.94 9.31 9.31 9.S8 ~9.31 9.58 Variance 7.84 6.4S 6.4S 20.70 6.4S 20.70 1 )bservadons 1 106 1 270 1 270 1 51 1 70 1 51 l | ' Test Results l Computed z value 2.00 -0.40 -0.40 P(Z<~) one-tail 0.02 0.34 0.34 z Critical one-tail 2.33 2.33 2.33 P(Z<=z) two-tail 0.01 0.17 0.17 z Critical two-ta~1 2.S8 2.S8 2.Sg HO: Cases are identical Accept Accept Accept The results of the difference in means test are summarized in Table 46. These tests support, for Me most part, hypotheses 3a, 3b, and 3c, that the directional movement affects saturation headway. Even for Pose tests Tat do not support the hypotheses, We relative values of Me saturation headways are as expected. Table 46. Summary of Hypothesis Tests ; , ................ ................ .................. .................... : :.~::~:~:::~:~:~:~:~:~: m::: :.:::: :::::::::::::: +.:.:.: :: :~:~:~.:::~:~: :.:: ::.:::::::::: ^:.::: :.:.: ^:~:~: :~: :~:::::::: :: ::::: all::: :~ you-= ;t test; ~ ~ ~ ~ ~ ~e ~-e e~ ~ ~ ~ - - -~90Ft ~ - ~e;;, . ., it. , . ,. ,. , , , .... ... , ..... , . Test 3a(~1~111) N Y N Y N Test 3b(Ll>RT) Y Y Y Y N Test 3c(TH>RT) Y N Y Y N Note: Y = supports hypothesis. N = does not support hypothesis. ,. Effect of Heavy Vehicles. The saturation headway data for each degree of conflict case are divided according to the vehicle Me of the subject vehicle. These data are shown in Table 47. The results of Me difference in means test are given in Table 48. In most cases, hypothesis test 4a is supported by Me data. ~ aD cases, the mean values for passenger cars are less than the mean values for light trucks or heavy trucks.

OCR for page 31
38 Table 47. Saturation Headway by Vehicle Type Passenger Car Mean Std Dev Observations 3.8S 1.37 3814 4.77 1.SS 1849 S.86 1.6S 1984 7.3S 2.01 1664 9.49 2.90 419 Light Truck Mean Std Dev Observations 5.S3 2.00 14 4.78 1.85 15 6.72 1.81 3 7.84 2.12 18 6.84 3.76 2 Heady Truck Mean Std Dev Observations S.S9 1.74 10 6.94 1.93 8.86 4.39 9 11.08 5.14 9 12.25 2.88 S Motorcycle Mean Std lees Observations 3.Sg 1.63 17 3.43 1.68 4 6.S0 1.94 11 6.90 0.93 g 7.42 N/A 1 Table 48. Hypothesis Tests. Vehicle Type Mean Mown Variance Observations I' 2;2'22 '' "' "' '' ':""'' "'"'"''I'' '' ~ '2 I"''"" "'" ''' ' ' ' ~ i.,. . ,~., , ,4 .. .... ,l ;,. . 1 T Trlc | PC | Trk | PC 7 Tlk | PC T Turk S.55 4.7g 4.87 5.89 S.29 7.38 6.7S 3.31 2.41 4.33 2.73 lS.S6 4.03 13.48 24 1849 21 1984 12 1664 27 l ) t O C 4.S8 -1.36 -2.16 -2.23 0.00 0.09 0.02 0.01 2.33 2.33 2.33 2.33 0.00 0.04 0.01 0.01 2.S8 2.S8 2.S8 2.Sg Re j ect Accept Accept Accept PC 3.85 1.88 3814 Hypothesized Mean Difference z P(Z<=z) onedail z Critical one-tail P(Z<-z~tw~tail z Critical two-tail Conclusion: Cases are Identical Not sufficient data for hypotheses tests Effect of Opposing and Conflicting Movements. It bus already been shown that the directional movement of Me subject vehicle affects its samraffon headway. To obtain further insights on the effect of directional movement, consideration was given to specific combinations of subject, opposing, and conflicting movements. AD nine movement combinations for case 2 are shown In Table 49. Movement pairs that either cross or merge have mean saturation headway values that range from 4.92 seconds to 6.29 seconds. The unweighted mean value of these means is 5.8 seconds. Movement pairs that do not cross or merge have saturation headway values Mat range from 4.15 to 4.78 seconds. The unweighted mean value for these means is 4.4 seconds. These measurements confirm that the degree of conflict between vehicles significantly affects saturation headway values. All 27 movement combinations for case 3 are shown in

OCR for page 31
39 Table 49. Movement pairs Mat either cross or merge have saturation headway values that range from 5.5 seconds to 7.4 seconds, wad an unweighed mean of the mean values of 6.3 seconds. Movement pairs Mat do not cross or merge have saturation headway values that range from 4.4 seconds to 5.7 seconds, with a mean of S.1 seconds. Again, these measurements confirm Me importance of Me specific level of degree of conflict between movements. Selected topped for case 4 also show the effect of degree of conflict. The two values with no conflicts are 5.3 seconds and 7.7 seconds. The two values wad conflicts between aD three movements are 8.3 and 8.S seconds. Figure 16 ~mmanzes these results for these three cases. The x ems values in this figure are defined as follows. The first number is the degree-of-conflict case. Y or N denotes if a conflict between the movements excess ~ or not (N). Table 49. Saturation Headways for Various Conflicting Vehicle Movements for Group 1 Sites 1 ' 2 2 222 ' 22 'I''' - ' '1 '' ' ' ' I'' '"''""'' ' '' ;':~''1"'' " "' '''''''''''''' -1"''''* ' - " ' ' "' '' ' ............... . . .. .. . . . ... . . . i . ... .. . . . . ~ | `~ 1 l ~ D CD~i~g ~DV~ Case2 All 4.78 1.S6 1M4 SLT~LT 4.92 1.76 19 SLOTH 5.70 1.61 lSO SLT~RT 5.78 1.83 34 STILT 6.16 1.55 181 STEALTH 4.54 1.41 1165 STEWART 4.78 1.63 166 SRT~LT 6.29 1.30 30 SRT~TH 4.22 l.S1 168 SRT~ORT 4.1S 1.42 29 Case3 All S.88 1.69 2007 SLT~LLT S.72 1.3S 39 SLT OCR for page 31
40 ,,10 .' 5 co .......... .......... 0 2 Degree of ConRict Figure 16. Effect of Turning Movement Conflict Combinations on Saturation Headway Table 50. Effect of Time in Queue on Saturation Headway 4.74 1.S7 S78 4.84 1.S3 S13 4.71 l.SO 217 4.77 1.37 82 4.47 1.70 42 4.26 1.40 29 Effect of Time in Queue. The effect of time in queue on Be saturation headway was not listed earlier as a specific hypothesis tom However, Me influence of time in queue of the subject vehicle has been suggest to adversely affect the critical gap and follow up time for TWSC intersections. ~ addition, Were was some concern that sites from this current sty with low volumes may result In lower pressure on the subject approach Diver Bus affecffng the comparability of saturation headway values between sites. The data were divided according to the time In queue of the subject approach vehicle. A review of the dam presented in Table 50 does not support any clear effect of Be time in queue on saturation headway. This is probably because at AWSC intersections, drivers are assured of getting through He mtersecdon; this contrasts And TWSC Intersections where this assurance is not guaranty. '1 2 ~ 2' ' ' ' '' ' ''I $ ~' ' - ~: ' ' ' 1 ~ Awe .................. .... ... ....... . .. .................. . .. ... . ... . L]O I Mean 3.80 StDev 1.36 Obs 1884 10-20 Mean 3.93 StDev 1.37 Obs 1011 20-30 Mean ;^ 3.9S StDev 1.4S Obs 483 3~40 Mean 4.02 StDev 1.31 Obs 2S0 40-50 Mean 4.03 StDev 1.15 Obs 136 >50 Mean StDev Obs .:~:-.:.:.:-.-.:.:.:~:-.:.:-:.:-:~:~:.:-:.:.:.:~.:-~:~.:-.:~:~:.:.:.:.: .:.:.:::~:-.:.:.:-.:-:::-:-:::::-~--~.:~:-:-:.:.:~:~.:.:-~:-:-:.:.:-.:. .:-.: :~:.:-:.--:~--.:.:.-.:::::: ~.:.:~:.:.--.:~.:-.-.:.:-:-:-:~:- . ~ ~ . ~. S.96 7.36 1.73 2.04 1142 997 S.88 7.48 1.68 2.07 485 441 S.62 7.2S 1.44 2.21 221 161 S.78 7.0S 1.60 1.61 104 66 S.87 7.S9 1.60 2.32 39 26 S.22 6.84 0.91 1.28 14 8 9.36 2.S2 248 9.90 3.49 104 9~6.0 2;S7 4S 9.66 4.65 21 8.03 2.33 8 7.36 N/A 1

OCR for page 31
41 HYPOTHESIS TESTIN~GROUP 2, 3, AND 4 SITES Group 2 Sites. Group 2 sites are those with single lanes on both the subject and opposing approaches and two lanes on one conflicting approach and one lane on He over conflicting approach. The saturation headways for the Group 2 sites are compared to the values measured at the Group ~ sites (single lanes on each approach) for comparable degree of conflict cases. These values are shown in Table 51. Hypothesis 2a proposed Hat He headways for Group ~ sites were less Man for Group 2 sites. The results of the difference in means tests shown In Tables 52 and 53, however, do not support this hypothesis in four of the five cases tests. That is, He saturation headways for Groups ~ and 2 sites are not stadstically different. - Table 52. Hypothesis Test 2 for Each Degree of Conflict Case Sacs Mean Variance Ions Z Test Remlts Computed z value P(Z<=z) one-tail z Critical one-tail P(Z<~) To z critical two-tail HO: Cases are identical Table S1. Headway Data for Groups 1 and 2 for Hypothesis 2a Tests _~~~ _ Case 1 Mean 3.86 4.00 StDev 1.38 1.30 Obs 38SS 249 Case 2 Mean 4.78 S.S2 SUDev 1.S6 1.79 Obs 1874 60 Case 3 Mean S.88 S.70 SUDev 1.69 1.47 Obs 2007 349 Case 4 Mean 7.37 7.34 SUDev 2.05 1.62 Obs 1699 295 . Case S Mean 9.S0 9.31 StDev 2.92 2.61 Obs 427 142 ... . ~'1''"''"': '''I'' ~ - ~; ;t 2'222' """'"' ""''I"""'''' ' '1'~""'-''1"' 4.00 3.86 S.S2 4.78 S.70 S.88 1.69 1.91 3.2 2.43 2.16 2.84 249 38SS 60 1874 349 2007 .~ 1. 58 1 3~18 1 -2.04 0.06 0.00 0.02 2.33 2.33 2.33 0.03 0.00 0.01 2.S8 2.Sg 2.SS Accept Rhea Accept

OCR for page 31
42 Table 53. Hypothesis Test 2 for Each Degree of Conflict Case . ..,.,., ., ., ., ., ., ., . . . . . . ... . . . . . . . . . . . . . . . .. . .. . .. ........ . . . .... . . . . . . . . . . . . . . . ... . ... . ... ...... ..... . ~ ~ ~ .. ... .. . .. . . . . . . . . . . . . . . . . . . . . . . . .. . . .. . . ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. .. . . . . . . . ....... .. , 2 ; Statistics Mean Variance Ob~ons 7.34 2.62 29S 7.37 4.20 1699 9.31 6.81 142 ALSO g.S2 427 Z Test Results Confuted z value P(Z<=z) one-tail z Critical one-tail P(Z<=z) tw~tail z Critical tw~tail HO: Cases are identical -0.26 0.40 2.33 0.20 2.58 Accept -0.74 0.23 2.33 0.11 2.S8 Accept Group 3 Sites. Group 3 sites have one subject approach lane, two opposing lanes, and either one or two conflicting approach lanes. AR Group 3 sites are T- intersections with two opposing approach lanes and only one conflicting approach. Group 3a sites have one conflicting approach lane; Group 3b sites have two conflicting approach lanes. Table 54 lists the saturation headways for Group 3, Group 3a, and Group 3b, for four of the five degree of conflict cases. The number of observations permit difference in means tests for degree of conflict cases ~ and 2 only. The results of these tests are given in Table 54. The first test determines if Me Group 3a and 3b sites yield similar or different saturation headway values. For degree of conflict cases ~ and 2, the saturation headway values are sign~ficandy different showing that the number of conflicting lanes affects the value of saturation headway. This difference is likely due to the increased distance traveled through the intersection. These are listed as tests 5a and 5b. Test 2a shows Tat We number of conflicting lanes does not affect Me saturation headway. Test 2b shows mixed results. Test 2b(~) shows Mat if Mere are bow 2 lanes on the opposing and conflicting approaches, Me saturation headway is greater Man for single lane sites. Just adding a lane on Me opposing approach, as shown in Test 2b(2), does not cause a significant difference, however. Table 54. Saturation Headways for Group 3 Case 1 Mean Stand Dev Observations 4.16 1.19 1707 4.13 1.18 1493 4.3S 1.20 214 Case 2 Mean Stand Dev Observations 4.7S 1.46 1226 4.71 1.42 1105 S.19 1.67 121 Case 3 Mean Stand Dev Observations 7.S2 2.63 12 6.22 1.47 202 Case 4 Mean Stand Dev Ob~v~ons 6.6S 1.05 7 7.35 2.15 212 Case 5 Mean Stand Dev Observations ibis case not applicable for this geometric group.

OCR for page 31
43 Table 55. Hypo~esis Tests for Groups 1, 2, and 3 :~ ::::::::::::::::::::::::::':':':':':':':':': :':':':':':':':':':':':' ':': :::':':'::: :::::':::::::::::1 ::::::: .................................................................................................................................................... : ~ ;~ StaffsUcs 1 1 1 1 1 1 Geometric Group 1 2 1 3b 2 3a Mean 3.86 4.01 3.86 4.3S 4.01 4.13 Vanance 1.91 1.63 1.91 1.4S 1.63 1.39 Obsavations 38SS 229 38SS 214 229 1493 _ Z T"t Remlts Computed z value -1.71 -S.71 -1.32 P(Z<=z) one-tail 0.04 0.00 0.09 z Ckitical one-tail 2.33 2.33 2.33 P(Z<=z) two~ail 0.02 0.00 O.OS z Critical two~ail 2.S8 2.S8 2.S8 HO: (~ are identi~t Accept Reject Ac~xpt Table 56. Hypothesis Tests for Groups 1, 2, and 3 'tadstics - 1 1 1 1 1 Geometric Group 3a-NC1 3~NC1 3a-NC2 3~NC2 Mean 4.3S 4.13 S.19 4.71 Variance 1.4S 1.39 2.8 2.03 Observations 214 1493 121 llOS Z Test Results Con~uted z vatue 2.SO 3.04 P(Z<=z) one-tail 0.01 0.00 z C\itical one-tail 2.33 2.33 P(Z<=z) two~ail 0.00 0.00 z Criticaltwo-tail 2.Sg 2.SS | HO: Cases are identical Reject Reject Group 4 Sites. Group 4 ~ncludes sites w~th a s~ngle lane on the subject approach, hvo lanes on the oppos~ng approach, and ei~er one or two lanes on ~e conflictug approaches. Stadshcal tesm ~ma~ in Tables 57, 58, and 59 show ~at the number of lanes on the conflic~ng approaches sig~ficandy affects the samradon headway. Table 57. Saturabon Headway Data for Group 4 Sites 22 ~.~ ., , ,,,. . .......................... , ......................................... .............................................................................. 1 3.72 .~.36 Regect 2 S.OS S.42 A~(ma~al) 3 S.60 6.17 Reject 4 7.19 7.77 R~ect S 9.43 lO.S7 Reject

OCR for page 31
44 Table 58. Hypothesis Tests; for Group 4 . ~. ~................................................... - ~........ , ... , ., ,.~ Staff~ffcs Geometric Group 4a 4b 4a 4b 4a 4b Mean 3.72 4.36 S.0S S.42 S.60 6.17 Variance 2.18 2.2 2.9S 3.00 2.24 3.31 Observations 99 121 S2 132 208 376 Z Test Rests Computed z value -3.16 -1.31 ~4.04 P(Z<=z) one-tail 0.00 0.10 0.00 z Critical one-tail 2.33 2.33 2.33 P(Z<=z) two-tail 0.00 0.0S 0.00 z Critical hro-tail 2.S8 2.S8 2.S8 He: Cases are identical Reject AT Reject Table 59. Hypothesis Tests for Group 4 Staff - cs Geometric Group 4a 4b 4a 4b Mean 7.19 7.77 9.43 10.S7 Variance 2.96 4.13 S.3S lS.05 Observations 1 236 1 439 1 114 1 197 Z Test Results Computed z value -3.92 -3.2S P(Z<=z) one-tail 0.00 0.00 z Critical one-tail 2.33 2.33 P(Z<=z) tw~tail 0.00 0.00 z Critical tw~tail 2.S8 2.Sg Ho: Cases are identical Reject Reject HYPOTHESIS TESTIN~GROUP 5 AND 6 SITES Group 5 Sites. Saturation headways for sites wad two lards on He subject approach are given in Tables 60 and 61. Table 60 shows He six sites USA two lanes on each of He four intersection approaches. Table 61 shows the data combined for all sites wad two lanes on the subject approach. These tables clearly show three points: The saturation headway increases wad degree of conflict. The saturation headway increases as the number of vehicles faced on the opposing and conflicting approaches increases. The variability of He saturation headway (as shown by He standard denadon) also increases as bow He degree of conflict and the number of vehicles faced increases. Group 6 Sites. Saturation headways for sites with three lanes on each approach are given in Table 62. These data clearly show three points: I: . .. . ~The saturation headway increases ninth degree of conflict. The saturation headway increases as the number of vehicles faced on tile opposing and conflicting approaches increases. The vanabili~ of He saturation headway (as shown by the standard deviation) also increases as both the degree of conflict and the number of vehicles faced increases.

OCR for page 31
45 Table 60. Saturation Headways, Group 5a Sites ...... a , ....... '''''' ''''" '' ' ' ""'"'"1 .... . , . 1 0 4.31 1.S3 S38 2 1 4.93 1.71 497 2 2+ 6.37 2.19 12S 3 1 6.44 1.91 609 3 2+ 7.00 1.90 76 4 2 7.13 1.87 1432 4 3 7.86 2.0S S86 4 4+ 8.87 2.SS 147 5 3 8.08 1.79 684 S 4 8.76 2.10 474 S 5 9.87 2.32 189 l S 6+ 10.87 3.24 81 Notes: Case is He degree of conflict case. NumVeh is Be number of opposing and conflicting vehicles faced by He subject driver. Table 61. Saturation Headways, All Group 5 Sites ................ ................. ............ ............ .. . ~............ ........ .......................... ........................ ............................................ , . ~,.,. . ~.......................................... ........................................... _ 0 4.32 1.49 781 2 1 5.01 1.69 6S7 2 2 6.27 2.1S 1S8 3 1 6.41 1.81 867 3 2 7.09 2.13 98 4 2 7.22 1.86 179S 4 3 7.98 2.07 717 4 4 9.16 2.73 167 S 3 8.32 1.87 81S S 4 9.09 2.44 S49 S S 10.14 2.S2 229 1 S 6 11.63 4.88 100 Notes: Case is He degree of conflict case. Num~eh is He number of opposing and conflicting vehicles faced by the subject driver. Table 62. Saturation Headways, Group 6 Sites ... ,2. ~, '.2,2.'. - ., ~ 1 0 4.45 1.S6 170 2 1 6.17 1.73 49 2 2 7.04 1.76 20 2 3 7.S1 1.6S S 3 1 6.81 1.38 146 3 2 7.47 1.86 46 3 3 ~7.94 2.02 11 4 2 8.26 1.81 96 4 3 8.78 1.72 48 4 4 9.70 1.92 32 4 S-8 12.S1 3.39 12 S 3 10.23 1.81 IS S 4 11.32 2.27 14 S S 11.6S 1.97 S S 6-9 13.43 3.01 14 Notes: Case is He degree of conflict case. NumVeh is He number of opposing and conflicting vehicles faced by He subject driver. REGRESSION ANALYSIS-GROllPS 14 SITES The analysis presented ~ the previous sections showed that the most significant factors affecting saturation headway were the mining movement of the subject vehicle, the degree of conflict fat by the subject vehicle, the vehicle type, and the geometry of the intersection. Regression analysis was used to determine ache effect of each of these variables collectively on saturation headway. The results of the analysis are presented In Table 63. Single lane sites are designated Group I. Group 2 sites have one lane on the opposing approach and two lanes on either of the conflicting approaches. Group 3 sites (T- intersections) have two lanes on the opposing approach and either one or two lanes on the conflicting approaches. Group 4 sites have at least two lanes on the opposing approaches and one or two lanes on the conflicting approaches. The details of the geometry of each site were given earlier in this report in Table 25. The constant term In each of the seven models presented represents the ideal situation, the saturation headway for passenger cars traveling straight through a single lane approach intersection with no opposing or conflicting vehicles at the intersection. When Group ~ sites are included in Me analysis, this base case saturation headway varies Tom 3.~S seconds to 3.93 seconds. This low variation indicates that this is a very stable value and can be used with confidence as a base value Dom which to compute other saturation headway values. AD constant terms are statistically different than zero. The effect of the turning movement is clearly indicated in the table. The sa~rabon headway for left turning vehicles is higher than for a Trough vehicle by 0.23 to 0.34 seconds. The saturation headway for right turning vehicles is less than for a through vehicle by 0.52 to 0.57 seconds. Bow sets of values are statistically significant. The saturation headway for heavy vehicles is higher than for passenger cars by 1.41 to 1.65 seconds.

OCR for page 31
46 The degree of conflict with opposing and conflicting vehicles affects the saturation headway. For case 2, the increase is from 0.84 to 0.91 seconds. For case 5, the increase is from 5.57 to 5.65 seconds. AD values are statistic ally s ignif~c ant. The effect of geometry is significant, and is particularly dependent on the number of conflicting lanes. For T- ~ntersections, increasing the number of opposing lanes Table 63. Regression Models for Saturation Headway Groups 1~ Sites Bom one to two increases the saturation headway by 0.12 seconds. Increasing the number of conflicting lanes Mom one to two increases the saturation headway by 0.42 seconds. For 4-leg intersections, increasing only the number of opposing^lanes from one to two does not significantly increase the saturation headway. But increasing the number of conflicting lanes, if the number of opposing lanes increases from one to two, does increase the saturation headway by 0.56 to 0.57 seconds. . ~ --; .. ......... I . ~ .............. ... ........... . ~.... . ~, ~. .... ...... .......... . 9,862 10,957 16,415 16,41S 0.668 0.676 0.688 0.6gS 0.446 0.457 0.473 0.473 0.446 0.457 0.473 0.473 1.716 1.716 1.703 1.703 ~'"''"''l.2 ."..'"2""1''''" ~ ~ , , .. , ., ~ ..... ~ . . 0~( ~1 3.88 I 0. 3 1 3.90 1 0 O ~1 3.93 1 0.02 O.OS 0.26 0.05 0.24 0.04 0.23 0.04 0.05 -0.52 0.05 -0.55 0.05 -O.S7 0.05 0.19 1.41 0.19 1.65 0.14 1.64 0.14 O.OS 0.93 0.05 0.84 0.04 0.8S 0.04 0.06 1.94 O.OS 1.88 0.04 l.gS 0.04 0.04 3.01 0.04 3.06 0.04 3.03 0.04 0.09 S.57 0.08 S.6S 0.06 S.63 0.06 0.12 0.04 0.42 0.07 0.40 0.07 0.00 0.07 . O.S7 O.OS ' O.S6 0 OS Observations 16,41S R R squared add R squared SEE 0.688 0.473 0.473 1.702 cot LT RT TRK Case2 Cases Case4 CaseS Group3A Group3B Group4A Gioup4B 3.88 0.34 -0.S4 1.44 0.91 2.02 2.96 S.S8 3.90 0.24 -O.SS 1.6S 0.84 1.88 3.06 S.6S 0.12 0.42 O.S7 0.03 0.04 0.05 0.14 0.04 0.04 0.04 0.06 0.04 0.07 _ O.OS l Notes: Colitis the regression coefficient, ~ ~ the standard error of Me coefficient estimate. ~ is the Standard error ofthe dependent vanable, Marion headway.

OCR for page 31
47 REGRESSION ANALYSI~GROUPS 5 AND 6 SITES The its of the regression analysis for sites with two or more lanes on the subject approach is given in Table 64. The base saturation headway (the constant in the regression equation) representing passenger cars traveling through the intersection facing no opposing or conflicting vehicles is nearly the same for Group 5 (4.33 see) and Group 6 (4.26 sect sites. These values are about 0.4 seconds higher than for the Group ~ sites, reflecting the additional time required to travel through a multi-lane site. The more complex the degree of conflict, the higher the saturation headway. The terms for the venous degree of conflict cases reflect this. Table 64. Regression Models for Saturation Headways~roups 5 and 6 Sites The saturation headway for heavy vehicles is I.6 seconds higher than for passenger cars, for the Group 5 sites. There were not sufficient heavy vehicles In the Group 6 Ma for a s~isticaDy-signiiicant estimate of this factor to be made. The saturation headway for left turn vehicles was 0.4 to 0.5 seconds higher than for passenger cars. For Group 5 sites, the saturation headway for right turning vehicles was 0.7 seconds less than for through passenger cars. No estimate could be made for Group 6 sites, since there were no right mining vehicles during penods of continuous queueing at these sites. .. ...... . ...... ...................................................................... ................................................. .............................................................. ........................................................................................................ .................................................................................. .................................................. Observati news 1 5438 1 1495 1 6933 ~683 1 683 1 R | 0.63 1 0.74 1 0.64 | 0.78 | 0.78 | R squared 0.40 O.S4 0.41 0.61 0.61 adjusted R squared 0.40 O.S4 0.41 0.61 0.60 SEE 1.90 2.0S 1.98 1.74 1.74 l ~< 1 65 ~1 1 ; . Constant 4.34 0.08 4.32 0.14 4.33 0.07 4.26 O.1S 4.26 LT O.S4 0.07 0.39 0.13 O.S4 0.06 0.42 0.14 0.42 RT -1.01 0.12 ~.37 0.17 -0.71 0.10 0.00 0.00 0.00 l Tmck 137 0.19 3.00 0.4S 1.61 0.18 0.30 0.80 Case 2-1 0.61 0.12 0.88 0.21 0.66 0.10 1.73 0.28 1.73 0.28 Case 2-2 2.03 ^0.19 1.50 ~p.38 41.91 0.17 2.S7 0.41 2.Sg 0.41 Case 2-3 3.17 0.79 3.17 0.79 Case 3-1 2.07 0.11 l.9S 0.18 2.03 0.10 2.37 0.20 2.37 0.20 Case 3-2 2.82 0.23 3.12 0.46 2.85 0.21 3.0S 0.29 3.06 0.29 Case 3-3 3.S2 O.S4 3.S2 O.S4 Case 4-2 2.70 0.10 3.16 0.17 2.79 0.09 3.82 0.22 3.82 0.22 Case 4-3 3.39 0.11 4.0S 0.22 3.S1 0.10 4.40 0.28 4.40 0.28 Case 4-4 4.38 0.18 6.87 0.48 4.66 0.17 S.31 0.34 S.32 034 Case 4-S 8.07 O.S2 8.07 O.S2 Case S-3 3.60 0.11 S.09 0.23 3.83 0.10 S.74 0.47 S.74 0.47 Case S-4 4.28 0.12 6.7S 0.27 4.61 0.11 6.84 0.48 6.84 0.48 Case S-S S.40 0.16 6.93 0.3S S.67 O.1S 7.13 0.79 7.13 0.79 l Case Sat 6.38 0.23 10.37 0.49 7.1S 0.21 8.97 0.49 8.97 0.48 0.1S 0.14 nnn Notes: Co OCR for page 31
48 BINDINGS AND RECOMMENDED VALUES This section of He report describes the analysis of saturation headway values for AWSC intersections, and the factors Hat affect these values. Otis ~nteresdng first to compare the results from this study with previous studies cited earlier in this report Table 65 shows these data for single lane (Group I) sites. While some of the specific values vary, the trend of increasing saturation headway values as the degree of convict increases is consistent for ad studies. Fable 65. Comparison Win Previous Studies of Saturation Headways , . ~e : : 1 ~1 4.0 1 3.9 I 1 3.21 3.0 1 3.9 1 3.9 : 2 ~2 1 5.6 I T 491 48 T 4-8 1 4.8 3 3 1 ~l 1 5.1 1 1 5.9 T 5-9 1 : 3 S.9 4 S.6 S.6 S.9 51 7.6 1 6.5 1 6.8 ~6.3 :71174 : 6 4 6.9 7.2 7.4 7.4 6.8 ~1 73 1 74 1 j1 8 1 5 1 9.0 1 8~4 T 8~0 T 95 1 95 1 95] Four hypotheses were proposed. The hypotheses are rest here and He findings and conclusions with respect to ache hypotheses are presented. Hypothesis ~ The saturation headway of a subject vehicle is dependent upon the;de~ee of conflict faced by the subject driver as meas~bytihe presence of vehicles on the opposing and conflicting approaches. Finding. The analysis of 12 single lane sites, including date bom45 separate approaches, shows that five separate cases are required to descnbe the degree of conflict experienced by a subject approach driver. These five cases are s~isticaDy distinct, forming a hierarchy of increasing values of saturation headway as the degree of conflict increases. This conclusion is support by both difference in means tests and regression analysis. Conclusion. This finding supports hypothesis I, including ad of the sub-hypo~eses la, Ib, Ic, and Id. Hypothesis 2 The saturation headway of the subject vehicle is dependent on intersection geometry, particularly He number of lanes on the conflicting approaches, the opposing approach, and the subject approach. Finding Creasing the number of lanes on the conings approaches, while maintaining a single lane on~the opposing approach does not significantly increase the saturation headway of the subject approach driver. This finding is based on a comparison of the five Group 2 sites untie the twelve Group ~ sites. It should be pointed out, however, Hat four of the five Group 2 sites had one lane on one conflicting approach and two lanes on the other conflicting approach, while one of the five sites had two lanes on both convicting approaches. It may be argued that two lanes on bow conflicting approaches for all sites would have resulted in a statistically significant difference between Group ~ and Group 2 sites, due to the increased have! distance for through vehicles and a greater potential for conflict between vehicles. This elect, however, was not found in the analysis. This finding is supported by bow difference in means tests and regression analysis.

OCR for page 31
49 Finding. Increasing the number of lanes on both the opposing and conflicting approaches increases the saturation headway of the subject approach vehicle. The analysis shows that there Is a significant difference between He headways measured at Group ~ sites and Group 3 and 4 sites. Finding Increasing the number of lanes on the subject, opposing, and conflicting approaches, as well as the number of vehicles on these approaches, increases the saturation headway of the subject approach vehicle. This finding is supported by regression analysis of Group 5 and Group 6 site data. Conclusion. These findings support Hypothesis 2. Hypothesis 3 The sa~rabon headway of the subject vehicle is dependent on its directional movement as well as the directional movement of He opposing and conflicting vehicles. Finding. The analysis of the data from the single lane approach sites show that the saturation headway of the subject approach Giver is affected by his or her directional movement This conclusion is based on both difference in means test and regression analysis. Finding The same analysis shows thatch saturation headway of He subject approach driver is affected by He specific combination of He directional movements of the subject, opposing, and conflicting approach vehicle. This finding is supported by a comparison of the saturation headway values for venous movement interaction combinations. Conclusion. These findings support Hypothesis 3, as well as the sub-hypo~eses 3a, 3b, 3c, and 3d. Hypothesis 4 The saturation headway ofthe subject vehicle is dependent on its vehicle type. Finding The analysis of the data for He single lane approach sites shows Hat the saturation headway of the subject approach Giver is a~ectedby the type of vehicle. This finding is supported by both difference In means test and regression analysis. Conclusion. This finding supports Hypothesis 4, particularly sub-hypothesis 4a. Table 66 shows the resulting saturation headway values for each geometry group and degree of conflict case, as well as adjustments for boning movement and vehicle type. Shaded ceils Ante table indicate parameters that are not applicable for a geometry group. Blank cells indicate values Cat were not available from the regression analysis. In general, He values presented In the table show a consistency w~thtihe hypotheses and He resuming findings and conclusions both within and between geometry groups. However, there are several cell values that are not consistent and should be re-assessed if a consistent set of values are to be used as He basis for a final cap acid model. Table 67 shows the final set of recommended values, including the following modifications of several of the regression results. These modifications, while minor, provide a consistency between all geometric groups and degree of conflict cases. The values for Group 4a sites have been Increased by 0. ~ seconds to be consistent with the data for Group 3a sites. The case ~ headways for Groups 5 and 6 have been increased by 0.2 seconds Stoat 4.3 to 4.5 seconds) to be consistent wig the Group 4b values (also 4.5 seconds). The Group 5, Case 4-2 headway was increased from7.1 seconds to 7.6 seconds to tee consistent ; with He Group 4b Case 4 value. The Group 5 Case 5-3 and 5-4 headways were increased EomS.2to 9.7 seconds end fiom 8.9 to ,. 9.7 seconds respectively to be consistent with the Group 4a value. The Group 6 left turn adjustment was Increased from 0.4 to 0.5 seconds to be consistent wig the Group 5 value. The Group 6 nghttum adjustment was set at -0.7 seconds to be consistent with the Group 5 value. The Groups 5 and 6 heavy vehicle adjuslments were set to I.7 seconds to be consistent with the values for the other groups.

OCR for page 31
50 Table 66. Saturation Headway Values from Regression Analysis 4.7 6 2 s.3 S 8 6 4 7.6 8.2 8.9 10.0 ll.S . ,- O.S -0.7 1.6 Notes: Case is the degree of conflict case. Bumped is the number of opposing and conflicting vehicles faced by the subject approach driver. IT and RT are the adjushnents forewarning vehicles. Harris the adjus~nent for heavy vehicles. - notes eases that are not applicable. Table 67. Recommended Saturation Headways for AWSC Intersections 3.9 4.7 9.6 4.0 4.8 ~ 9 7.1 3.9 (4.0) 5.1 7.4 4.7 (4~8) 5.8 (5.9) 7.0 (7~1) 4.3 (4~5) 5.3 6.4 7.6 5.0 6.2 7.1 (7.6) 7.8 9.0 4.3 (4~5) 6.0 6.8 7.4 8.1 8.7 9.6 12.3 9.7 10.0 9.6 (9.7) ........................ . G - , ., , , .. ,, , 0.2 -0.6 1.7 10.2 8.2 (9.7) 8.9 (9~7) 10.0 11 5 10.0 11.1 11.4 13 3 0.2 -0.6 1.7 0.5 -0.7 1.6 (1.7) 0.4 (0~5) (-o.7) (1.~ Notes: Case is the degree of conflict case. NumVeh is the numb" of opposing and conflicting vehicles faced by the suyoct approach driver. LT end RT are the adjushnents for turning vehicles. HVis the adjustment for heavy vehicles. - notes cases that are not applicable. The values fiom regression analysis are shown first in the cell; adjacent values in the same cell in pareses are the adjusted and recommended values.