overtaking and passing requirements as...

HIGHW.~~

~' :~''!

~'roceedings

.~i~°hteent~i Annual Meeting°

,-.

OVERTAKING AND PASSING REQUIREMENTS AS DETERMINED

FROM A MOVING VEHICLE

BY T. M. MATSON AND T. W. FoxsES,l Research Associates,

Yale University Bureau for Street Tra,~c Research

SYNOPSIS

By photographic methods observers in a moving car were able to measurethe times required by cars overtaking them to pass under various conditions.

It was found that drivers varied widely in the speeds, distances, and timeswhich they needed to overtake and pass another vehicle. Distance needed in-

creased with speed. At 30 miles per hour 650 ft. were required and at 50 milesper hour 1,050 ft. were required by the average driver (median values). In orderto take in 80 per cent of the drivers approximately k,200 ft. would be requiredat 50 miles per hour for an accelerative type of pass.The total time required to overtake and pass also increased somewhat as

speed increased. Thus, at 30 miles per hour the average driver took 8~ sec. andat 50 miles per hour 9% sec. to complete an accelerative type of pass (medianvalues). The time varied from a minimum of 6 sec. to a maximum of 20 sec. inindividual cases at speeds from 30 to 50 miles per hour.From the safety point of view it is of interest to see what clearances were

allowed by the passing drivers both to the rear and to the front of the vehiclebeing passed. The clearance to the rear showed a median value of 45 ft. at 50miles per hour in the accelerative type of pass, and the clearance to the frontshowed a median value of about 60 ft. However, some drivers approached asclose as 10 ft. to the rear and cut in as close as 30 ft. in front of the overtakenvehicle at 50 miles per hour.The clearance distances were found to be shorter in the New England area,

where crooked and hilly roads are the rule, than they were in the open countryof the west and midwest, where fairly long clear distances predominate. Thedrivers in the first area were apparently induced to pass under conditions whichwere considered hazardous by the drivers in the areas where longer clear dis-tances were frequently available. Thus the importance of providing sufficientclear sight distance on highways is demonstrated.

Those who are charged with the re-sponsibility of designing new highways orredesigning for the improvement ofalignment on e~usting highways are es-pecially concerned with the requirementsof overtaking and passing manoeuverssince such measurements furnish a meas-ure of adequate sight distance. Sincethe great bulk of trafric operation iscarried out on 2-lane highways, the prob-lem of adequate passing sight distancebecomes especially important in thedesign of the alignment features of theseroads.

PREVIOU6 STUDIES

Perhaps the first definite steps taken inrecognition of the hazards attendant to

i Bryant R. Burkhard assisted with thephotography.

overtaking and passing were the draftingand adoption of regulations by variousstate legislatures which prohibited motor-ists from overtaking and passing unlesscertain clear paths were available at thetime and place when such manoeuverswere carried out. Unquestionably, fac-tual data concerning the requirementsfor such manoeuvers were lacking forthe preparation of such regulations.This is manifest by a review of vehiclelaws in the various states which show arange of prohibition varying from 150 ft.in Arizona to as much as 1000 ft. inWisconsin.Other investigators became interested

in this problem as early as 1934 and madeestimates of the time and distance re-quirements for overtaking and passing

loo

ETERMINED

:aces,

e to measureis conditions.es, and timesce needed in-id at 50 mileses). In orderd be required

somewhat asok 8a sec. andpass (mediann of 20 sec. in

;arances wereof the vehicleof 45 ft. at 50z to the frontapproached as'she overtaken

England area,open countryurinate. Theditions whichger clear dis-3ing sufficient

g were the draftingnations by varioush prohibited motor-and passing unless're available at thei such manoeuversnquestionably, fac-the requirementswere lacking forsuch regulations.

~ review of vehicleCates which show aarying from 150 ft.uch as 1000 ft. in

became interestedy as 1934 and madee and distance re-,aking and passing

MATSON AND FORBES—OVERTAKING AND PASSING 101

(1) (2) (3)? The results arrived at bythese previous investigations did notaccurately reflect the actual performanceof the public and their vehicles under theusual circumstances of operation uponthe public ways. In some, certain as-sumptions were made on driver andvehicle performance; in others, only arelatively few vehicles and drivers actingunder specific instructions were involved,while in yet others the complete manoeu-ver was deduced from a large number ofactual cases wherein only a small pro-portion of the entire manoeuver wasrecorded, while still others arrived at theirconclusions on a completely a priori orcomputed basis.

vehicle under observation. Distancesto the rear and front of the overtakenvehicle at the beginning and end of themanoeuver were obtained by photo-graphic adaptation of stadia methods.(See Fig. 1.) A rapid, serial, snapshotcamera was used for recording the dis-tances projected by the tread dimensionsof the overtaking vehicle. Because ofthe approximate uniformity of tread di-mensions, errors which arose were negli-gible insofar as the total distance requiredfor overtaking and passing was concerned.The distance through which the over-taken vehicle moved during the manoeu-ver was obtained by calibrated speedom-eter and stop watch. Other pertinent

-~ -̂-~-~ r--- =1 ~-~C -~ M~ ~ -~ Y~ ~L-_-4=_i -1~ =='

B►CKSIOY~T~ DIJIAIICt D15T4NCE TR 4VtLL[D DURING ORSJ tORt S1GUi DIJT ANGE(DY GAM[RA) (6Y Ji0DW4iC 11 AND SPEEDOM[TtR~ (BT C4MERA)

OVtRT 4KIXG AND PASSING D13SANGE AS OBSERVED

Figure is Technique for Measuring Overtaking and Passing Distances

BASIS OF PRESENT INVESTIGATIONS

It was felt that the best way in whichovertaking and passing requirementscould be determined was by the measure-ment of the actual performance ofdrivers found operating on the highwaysince, admittedly, there is a wide varia-tion among drivers and vehicles.A new method of investigation was

developed, therefore, which could beapplied on the open highway undernormal driving conditions. This hasbeen described in detail previously (4).Observations were made from a movingvehicle which was overtaken by the

Q Figurea in parentheses refer to list ofreferences at end.

data were recorded by observations en-tered on a prepared form. Approxi-mately 800 observations were made ofas many different drivers and vehicleson highways throughout the New Eng-land, mid west, and far west areas.

ANALYSIS INTO TYPES OF PA88E8

There are two major types of overtak-ing and passing manoeuvers which maybe derived naturally from conditionsunder which the passes were made. Inthe first class, will be found the "flying"type of manoeuver, wherein the over-taking vehicle proceeds at constant ornearly constant speed so as to completethe entire pass without slowing down.In the second class, will be found the

roeDESIGN

type of manoeuver which may be de-fined as the "accelerative" type, whereinthe overtaking vehicle is following be-hind the overtaken vehicle and by ac-celeration increases its speed so as tocomplete the pass.In both the flying and accelerative

types of manoeuver a further classifi-cation arises from the manner in whichthe overtaking vehicle returns to the

Analysis of the data showed that thespeed of the overtaken vehicle was acontrolling element in practically allvalues of time and distance. Hence, itbecame necessary to distribute the ob-servations by speed classes ranging byten mile intervals from 10 to 50 milesper hour. These were the speeds of theovertaken vehicle and should not be

Figure 2. Flying Pass with Forced Return.Top—Start of Pass, View to Rear from Obser-vation Vehicle. Bottom—Completion of Pass,View to Front Showing Forced Return Dueto Counter Traffic.

right half of the roadway. Due to on-coming traffic or short visibility distancethe return may be hurried or "forced."Accordingly, both the flying and accelera-tive types have been further subdividedinto forced return and voluntary return.The general method which was followed

in the analysis of the data classified allpasses into these four types; 1, flying-forced (Fig. 2); 2, flying-voluntary; 3,accelerative-forced; 4, accelerative-volun-tary (Fig. 3).

Figure 3. Accelerative Pass With Volun-tary Return. Top—Start of Pass (View torear of observation vehicle, showing overtakingvehicle beginning to accelerate after waitingfor counter traffic). Bottom—Completion ofPass. (View to front showing free return.)

confused with the speed of the overtakingvehicle.It further developed that in each speed

range for each type of pass there is arather wide distribution of values. Ac-cordingly, the data are shown in the formof frequency distributions in order tointerpret better the values which wereobtained. (See Fig. 4.)Because of the wide range of values

and because of the assymetry and lack of

.a showed that theyen vehicle was ain practically allistance. Hence, itdistribute the ob-classes ranging by~m 10 to 50 milese the speeds of theid should not be

ve Pass With Volun-rt of Pass (View tole, showing overtakingTelerate after waitingatom—Completion ofowing free return.)

of the overtaking

i that in each speedof pass there is a

'ion of values. Ac-e shown in the formations in order tovalues which were4.)de range of values~ymetry and lack of


continuity found in the distributions, themedian was selected for comparativepurposes rather than the arithmeticmean. It is generally recognized thatthe median is preferable where suchcharacteristics prevail.

SPEED DISTRIBUTIONS

A total of 795 completed passes wereobserved. The speed distribution ofthese observations, as a whole, was

FORcEO RETURN

investigation. This summary is shownin Figures 4 and 5. It will be noted thatthe median values of overtaking andpassing distances fall approximately on astraight line. The values for 30 and 50miles per hour are given in Table 1.Thus, it is seen that there is only a slightdifference between the distances requiredfor the accelerative and flying passes ofthe voluntary return type. Contraryto expectation, the flying type requires

!/OLUNTARY RETURN■~~~■■~~r~~■■nip■~~■■~■■■n~►~~~■■

■■■t!■■~■■It■■■1■■t■■■f~!~~II■■■■■

C~e■■~I~~~II■t■■t■■t■t!■1■%!t■■■■t■■■■Cii:~~i■~~■■■1!t■■I■I!■!~!■t■■■■I~~■II■■■■■~~■■■■~~■II■■■t■■t

■■~■■r.■ ~■■■■■m■■■■~~~r ■■■■■■n■■t■■~i~I■■~~~■■■■t■■t■G~■■t■■■■■■t

1 ■■/II ~■■t ■■■■■■■ ~~~~1I1 ■■t■■■t■■■■■V■~I~■■■1■■VV■~!~~C~!■1■11■■ ►~■

visrvnce ~n reer

Figure 4. Distance Required to Overtake and Pass, Flying Types, All Areas, All Drivers

comparable to the usual curve of speeddistribution on state highways. Themodal value was 40 miles per hour, andthe percentage of observations falling atother speeds was nearly congruous withthe distribution of speed values shownfor a single state in a recent study (5)of speeds on Connecticut highways.

GENERAL SUMMATION OF OVERALL

DISTANCES

All passes of passenger vehicles weredistributed into the four basic typeswhich included both male and femaledrivers in the three geographic areas of

slightly longer distance. A similar situa-tion holds for the forced passes, but thedifference decreases at higher speeds.

Attention is further called to the factthat the 80 percentile value of the volun-tary type of manoeuver will provide forthe large majority of forced passes inboth the accelerative and flying types.The use of the 80 percentile values arosefrom an analysis of clearance timesallowed by drivers (6). In this analysisit was found that if a clearance between avehicle completing the pass and an on-coming vehicle was placed as low as isec. as a common sense minimum safe

104 DESIGN

clearance, approximately 20 per cent ofthe drivers who attempted a pass in theface of oncoming traffic operated withless than this time clearance value. Infact, 10 per cent or half of these operatedwith no measurable time clearance or inother words actually forced the oncomingdriver to give way on his half of theroadway. This left only 10 per cent whooperated with a clearance of 1 sec. orless, and the 80 per cent who operatedwithin apparently reasonable judgment

h

distance required by 80 per cent ofdrivers at any given speed. Inasmuchas these 80 percentile values practicallyenveloped the large majority of forcedpasses in both categories, the practica-bility of 80 percentile distances is demon-strated.

OVERALL DISTANCES—MALE DRIVER$

Further analysis of distances and timevalues seemed desirable and in order toeliminate as many variables as possible,

FORCED RETURN ✓OLUNTfIRYRETURN

Figure 5. Distance Required to Overtake and Pass, Accelerative Type, All Areas, All Drivers

showed a time clearance value of morethan 1 sec.In parallel fashion it seemed in order

to determine the overtaking and passing

TABLE 1

COMPARISON OF OVERALL DISTANCES*

(All Drivers—All Areas—Passenger Vehicles)

Speed inFlying Passes Accelerative Passes

M.P.H.Forced Voluntary Forced Voluntary

!t. .fc. re. ,ft.

30 550 650 450 600so 7soios

o7soi000

* Median Values

the following analyses have to deal withmale drivers and passenger vehicles only.The distance required to overtake andpass for all areas has been graphically setforth in Figures 6 and 7 according tocategories hereinbefore described. Valuesfor both types of passes for 30 and 50miles per hour are given in Table 2.It is shown that the median value of the

flying-voluntary type was greater thanthat of the forced return type by about175 to 400 ft. at the speeds compared.It is seen also that in the accelerativetype of pass the voluntary return typerequired from 100 to 200 ft. more dis-

80 per cent ofspeed. Inasmuchvalues practicallyMajority of forcedies, the practica-istances is demon-

-MALE DRIVER6

istances and timeand in order to

fables as possible,

1 Areas, All Drivers

have to deal withiger vehicles only.to overtake andyen graphically setd 7 according todescribed. Values.es for 30 and 50n in Table 2.iedian value of thewas greater thanrn type by aboutspeeds compared.i the accelerativeztary return type200 ft. more dis-

MATSON AND FORBES—OVERTAKING ~1ND PASSING 105

tance than the forced return type on the tendency, however, for the median valuesaverage. to increase slightly when female drivers

FORCEO RETUR/V YOLUNT.9RY RETURN

0~~a

~ o

.v

m

0

O N

O

N

ho

Di3fc~nce in feetFigure 6. Distance Required to Overtake and Pass, Flying Type, Males, All Areas

O~ ~Q

.vCo,

FORCED RETURN VOGUNTARy RETURN

Distgr~ce !n feed

Figure 7. Distance Required to Overtake and Pass, Accelerative Type, Males, All Areas

Comparison of these results with those are included. Hence, it is reasonable tofor all drivers shows that practically the deduce from these differences that thesame values were obtained. There is a female drivers as a group required slightly

106 DESIGN

more distance for overtaking and passingthan the males.

INFLUENCE OF GEOGRAPHICAL AREA

It was noted in flat terrain wheregreater sight distances prevail that thenatural driving habits of motorists re-sulted in greater distances for overtakingand passing. Accordingly, all data formale drivers were tabulated into twocategories so as to compare the observa-tions made in the hilly terrain of NewEngland with those obtained in the flattersections of the middle and far westernstates. These are set forth in Table 3.It will be noted that the flying-volun-

tary type of manoeuver observed in themiddle and far western areas showed

TABLE 2

COMPARISON OF OVERALL DISTANCES*

(Male Drivers Only—All Areas—PassengerVehicles)

Speed inFlying Passes Accelerative Passes


,fc. Jt. !t. Jt.

30 500 675 450 55050 700 1100 ...750 950

*Median Values

i

from 100 to 425 ft. greater distance thanthose for New England. The accelera-tive-voluntary classification showed arange of 200 to 250 ft. greater distancein the middle and far western areas than

~ were found in New England areas.Since the number of cases for the fiying-forced type when separated by areas wastoo small to be highly significant, it wasomitted from the table. For the accel-erative-forced type, median distances

~~ showed a relationship similar to thevoluntary types.Thus, it is seen that when there is

prevailing limited sight distances andI drivers realize that more adequate sight

distances do not occur frequently, theywill execute passes under strain in face of

possible hazard, which would ordinarilybe refused where longer sight distancesare generally to be found. This inter-pretation is borne out by reference to theaccelerative passes under forced returnconditions and also by clearance timespreviously reported for the two areas.(Op. cit. (6).) If this interpretation iscorrect, the 80 percentile value from thevoluntary conditions should indicate thesight distance which is adequate to in-clude the forced passes under conditionswhich are generally corr~fortable. (Notethe foregoing discussion of &0 percentilevalues and forced passes.) Thus, while

TABLE 3

DISTANCES REQUIRED TO OVERTAKE AND PASS*

(Males—Passenger Cars Only)

Flying-VoluntaryAccelerative-Voluntary

Speed inM.P.H.

New Eng. Mid &Far New Eng. Mid a~ FarAreas West Areas West

ft. ft. ft. ft.

30 650 750 500 70040 750 850 650 80050 825 1250 800 1050

Flying-Forced Accelerative-Forced

30 ... ... 450 50040 ... ... 650 67550 ... ... 675 775

* Median Values.

passes are safely completed under shortersight distances., the hazards attendantthereto are unquestionably higher.

CLEARANCE DISTANCES—REAR

In analyzing the component parts ofthe total distance required to overtakeand pass, it is not without interest tonote the range of clearance distancesallowed by the driving public both to therear and the front of the overtaken ve-hicle. A general summary of all dis-stances which were obtained is set forthgraphically in Figure 8, Analysis of thedata showed that the clearance distance

would ordinarily'r sight distancesund. This inter-~y reference to theder forced return~ clearance timesr the two areas.interpretation is

le value from thecould indicate theadequate to in-under conditions

nfortable. (Noten of 80 percentile'.es.) Thus, while

OVERTAKE AND PASS*

•Cars Only)

Accelerative-Voluntary

gar New Eng. Mid ~ FarAreas West

ft. Jt.

500 700650 800800 1050

Accelerative-Forced

450 500650 675675 775

eted under shorteriazards attendantably higher.

NCES—REAR

mponent parts ofuired to overtakeithout interest toearance distancespublic both to thethe overtaken ve-mary of all dis-tained is set forth

Analysis of theclearance distance


to the rear of the overtaken vehicle wasnot influenced by the type of return. Inthe accelerative type regardless of thetype of return the median values ofdistances to the rear of the overtakenvehicle ranged from 25 ft. at 10 miles perhour to 45 ft. at 50 miles per hour. Inthe flying type of manoeuver the dis-tances to the rear of the overtakenvehicle were exceptionally constant forall speed classes, being approximately75 ft. It should be noted that these aremedian values and that considerable

type of pass varied linearly with speedfrom 75 ft. at 30 miles per hour to 65 ft.at 50 miles per hour. Note in thisinstance that the clearance distancedecreased with increase of speed. Thisdecrease was probably due to lack ofaccuracy in judgment at higher speedswhere oncoming traffic was present andhigher relative velocities at lower speedsof overtaking. For the flying-voluntarytype the clearance distance to the frontshowed a practically constant medianvalue from 10(~ to 120 ft. at all ranges of

ACCtLERATIVE TYPt

Z ~f a ~ ~z ~

io

'~ z ~ ~ ~ ~

~p e a o 5o c7 8'J tloOVER~ANtN Yl41C~E

DIST ANCt TO 0.t►q ~ DIgTANCE i0 F40NT

FLYING TVPt ~

~ ai f

~ ~ ~ ~ `_`z ~ ~ ~ ~ ~ ~y

p ~o W oo c o

(OVERTAKEN YEpiClt I 1L

iOi57►NGE TO 0.tA0. DIJTANCt i0 FRONT

Figure 8. Clearance Distance—Feet

deviation from them occurred. Thus,in the accelerative type the clearancedistance to the rear of. the overtakenvehicle ranged from as low as 10 ft. to asmuch as 90 ft. Similar values in theflying type of manoeuver ranged from20 ft. to 150 ft.

CLEARANCE DISTANCES—FORWARD

The clearance distances allowed infront of the overtaken vehicle show awide range of distribution. Medianvalues of clearance to the front of theovertaken vehicle for the flying-forced

speed, thus paralleling the constant rearclearance mentioned above.

Special significance accrues to the clear-ance distances allowed in front of theovertaken vehicle under accelerative-forced conditions. Clearance values ob-tained for this type of manoeuver havebeen set forth graphically in Figure 9.It will be noted therefrom that themedian value of forced return was fairlyconstant at all speeds at about 60 ft.Isolated values, however, range from 30to 150 ft. For purposes of comparisonthe 80 percentile values of the voluntary

108 DESIGN

type of return are also set forth. This

shows a fairly constant value of about

110 ft. As was the case for the total

distance, the 80 percentile value is suffi-

cient to include nearly all of the forced

return type. Here again, it is to be

noted that 80 per cent of the motorists

voluntarily performed in such a manner

as to include practically 100 per cent

under forced conditions.

~hQ

.0 V

d

~ ~

OftJ

0 N

dd

O

FORGED RETURN

pass is distributed from 5 to 14 sec.

Even under the most exacting condition,

(the accelerative-forced type of manoeu-ver) individual time values were distrib-uted from 6 sec. to as much as 16 sec.,whereas in the voluntary return type thetime required ranged from 6 to 19 sec.

Median values are given in Table 4.

Thus, the median times for forced

passes vary from 8.0 to 10.5 sec. at the

VOL UNTHRY RETURN

Disfonce in beef

Figure 9. Clearance Distances in Front of Overtaken Vehicles, Accelerative Type

TOTAL TIME REQUIRED TO OVERTAKE

AND PASS

While the total distance required to

overtake and pass is probably more

significant, the total time required to

complete a pass is of particular interest.

This holds especially on two lane road-

ways inasmuch as the entire time spent

by the overtaking vehicle in the left hand

lane is a time during which extreme

hazards are manifest.Accordingly, the times required to

overtake and pass have been analyzed

and shown graphically in Figures 10 and11. In the flying-forced type of manoeu-

ver the time required to overtake and

speeds compared, and those for the flying

are slightly longer than for the accelera-

tive. Again, attention is called to the

use of 80 percentile values for design

purposes due to the wide range of individ-ual values found.

TIME REQUIRED TO OVERTAKE

Further analysis was made of the time

reauired for the overtake alone, -that is,the time spent by the overtaking driver

from the instant he began his manoeuveruntil the foremost part of his vehicle was

opposite the mid point of the overtaken

vehicle. Figure 12 shows that for the

flying pass both forced and voluntary

110 DESIGN

TABLE 4COMPARISON OF TOTAL TIMES*

(Male Drivers Only—All Areas—PassengerVehicles)

Speed in Flying Passes Accelerative Passes


ft. ,ft. It. Jt.

30 8.0 10.5 S.0 10.050 10.5 12.0 9.5 11.5

* Median Values.

/5

~ io~y 6ac

60

V~4o~ zo

30p 20

Q "' io

~s

ON~~a ~'dQ /5

h o ios

ToTFIL fICCELERAT/VE

ter.:

sampling all types of drivers and vehicles.It is felt that this method reflects withsufficient accuracy the variations indriver and vehicle performance, and ashas been noted in all of the results, thereis a considerable range of individualperformance with regard to both timeand distance values at any observedspeed of operation.Our previous theoretical calculations

based on a reasonable theory of over-

N= 77

TOTAL FLY/Nd

Time in seconds

Figure 12. Time Required to Overtake, Forced sad Voluntary Returns Together

30 miles per hour approximately 32 sec.were required and at 50 miles per hourapproximately 42 sec. were required.

DI6CUSSION OF RESULTS

General Considerations. The resultswhich have been obtained lead inevi-tably to a general conclusion that thereis a wide difference in the operation ofautomobiles insofar as overtaking andpassing distances and times are con-cerned. It must be remembered thatthe results which have been obtained arebased on a new method of measuringovertaking and passing distances by

taking and passing and common senseestimates showed that the flying type ofpass if executed in the shortest distanceallowed by physical limitations anddriving comfort would be shorter thanpasses of the accelerative type. It issignificant to note, however, in actualoperation that this was not the case andthat generally speaking the accelerativetype of manoeuver required slightly lessdistance and time. It is to be concluded,therefore, that the average driver doesnot approach the minimum values whichare possible of achievement in the flyingpass. This is accounted for largely by

the g~front falthouhigherThe

and pftype cand fl;tary rdiffere250 fiaccele:flyingDes

result;the cbecoxnIt 1S TI

pasSindesignintim~speedtherethe 8~accelepractiinclucfelt ttthe dithatmotorsucce~will rto thaestab:after.it seethatprovevaluemansBringmentperceof fisignilBasegathEdista

vers and vehicles.hod reflects withe variations inormance, and asthe results, there;e of individualrd to both time~t any observed

tical calculationstheory of over-

as Together

x common sensehe flying type ofshortest distancelimitations andbe shorter than,ive type. It is~vever, in actualnot the case andthe accelerative

sired slightly lessto be concluded,rage driver doesum values whichent in the flyingd for largely by


the greater, clearance distances to thefront and rear in the flying types of pass,although they tended to be executed at ahigher average speed.The total distance required to overtake

and pass was considerably affected by thetype of return. Both in the accelerativeand flying type of manoeuver the volun-tary return utilized more distance, thedifference of median values amounting to250 ft. at 50 miles per hour for theaccelerative types and 300 ft. for theflying types.

Design Speed. Without exception allresults for all types of passes show thatthe overtaking and passing distancebecomes greater for higher speeds. Thus,it is readily seen that the overtaking andpassing distance which is chosen indesigning for modern traffic speeds isintimately bound up with the designspeed of the roadway. In connectiontherewith, it is significant to note thatthe 80 percentile distance found in theaccelerative-voluntary type includedpractically all passes of the forced types,including the flying-forced type. It isfelt that this is especially important fromthe design point of view for it is believedthat whenever .there is any doubt in amotorist's mind as to whether he cansuccessfully complete a pass or not, hewill reduce speed to a point comparableto that of the vehicle ahead and thereforeestablish any pass which he makes there-after as the accelerative type. Moreover,it seems from the design point of viewthat the fact that the accelerative typeproves shorter makes it desirable to usevalues obtained from that type ofmanoeuver when discussing or consid-ering overtaking and passing require-ments. In view of the fact that the 80percentile provides for the large majorityof forced passes, it seems especiallysignificant in considering design values.Based on the data which have beengathered, the overtaking and passingdistance, which includes 80 per cent

of all accelerative-voluntary passes,amounts to approximately 1150 ft. at50 miles per hour, 950 ft. at 40 milesper hour, and 750 ft. at 30 miles perhour.

Areas. The results obtained for NewEngland and the more open areas of themiddle and far western sections showthat under prevailing limited sight dis-tances motorists will pass under con-ditions which would be consideredhazardous by motorists travelling underconditions of prevailingly greater sightdistance.

Clearances. Consideration of the me-dian clearance distances allowed in theflying types of pass showed that theaverage motorist allowed himself approx-imately 70 ft. headway before starting amanoeuver. In the accelerative typesthe motorist allowed himself from 25 ft.at 10 miles per hour to 45 ft. at 50 milesper hour. These latter values seemabsurdly low from the safety viewpointand reflect the questionable amount ofcaution which is exercised under theconditions attendant to the accelerative-forced type of pass.On the other hand, the average motor-

ist when forced to return to the right halfof the roadway in front of the overtakenvehicle allowed an average clearancedistance of about 60 ft. This clearancevalue was increased to an average of110 ft. under voluntary return conditions.The differences in drivers under varyingconditions is especially indicated in theseclearance values inasmuch as isolatedcases showed a range of distribution fromas low as 10 ft. to the rear to a minimumof 30 ft. to the front of the overtakenvehicle.

Times. For purposes of computingsight distances the amount of time re-quired for overtaking and passim seemsespecially pertinent. The results showthat the total time required to overtakeand pass increased with speed. For thelower speed of 30 miles per hour the

112 DESIGN

accelerative-forced type required 82 sec.to complete a pass. In the accelerative-voluntary type the median value of thetime required at 50 miles per houramounts to 112 sec. To this value of thetime required to overtake and pass mustbe added some additional time clearancevalues for the computation of sightdistance. In a previous paper (Op. cit.(6)) it was shown that 80 per cent of thedrivers allowed a clearance of 1 sec. ormore when passing in the face of oncom-ing traffic. It is beyond the scope ofthis paper to discuss the various otherfactors involved in the computation of asafe passing sight distance.

SUMMARY

1. Actual distances and times requiredto overtake and pass by the driving publicwere measured. A total of 795 observa-tions were obtained in New England,mid western and far western states.

2. A photographic method was devel-oped for use from a moving vehicle.

3. Four types of passes were developedfor the purpose of analysis.

4. Distances and times increased lin-early with speed. At 50 miles per hourdrivers executing a "following" passrequired a median distance of 1000 feetand 80 per cent of them passed in lessthan 1150 ft.

5. The corresponding median timevalue was 11.5 sec.

6. The 80 percentile values provideadequate distance for practically all

observed passes under pressure. It is,therefore, suggested for design purposes.

7. Conditions of terrain apparentlyinfluenced driver behavior. Clearancesto rear and to front have been analyzed,together with corresponding time values.8. The results emphasize the fact that

traffic studies of this sort involve theinherent variability of the human factoras represented by the performance of thedriving public. The wide range of valuesobtained necessitates the simpler type ofstatistical treatment developed for theanalysis of biological and human data.Failure to recognize this range andvariability leads to conflicting and erro-neous interpretations.

REFERENCES

1. Massachusetts Highway Accident Survey.1934. W.P.A. of Massachusetts.

2. Dickinson, H. C., Highway Research Ab-stracGs, Oct. 1934.

3. Greenshields, B. D., "Distance and TimeRequirements to Overtake and PassCars," Proceedings Highway ResearchBoard, Vol. 15, pp. 332-342, 1935.

4. Matson, T. M. and Forbes, T. W., "A Pre-liminary Report on the Measurement ofOvertaking and Passing Distances," Pro-ceedings Inst. Traffic Engineers, Oct. 1938.

5. Tilden, C. J., Morris, D. L., Martin, T. M.C., and Russell, E. W., "Motor VehicleSpeeds on Connecticut Highways," Com-mittee on Transportation, Yale Univer-sity, 1936.

6. Forbes, T. W. and Matson, T. M., "DriverJudgments in Overtaking and Passing,"Paper before Amer. Psychol. Asan. Meet-ing, Columbus, Ohio, 1938.

DISCUSSION ON OVERTAKING AND PASSING REQUIREMENTS

DR. B. D. GREENSHIELDS~ College of theCtity of New York: It is gratifying to knowthat the authors' results agree so closelywith those I arrived at by an entirelydifferent method. (Proceedings, HighwayResearch Board, Vol. 15, pp. 332-342.)

', Of significance is the fact that the time'~ I found was the clearance time required

on the opposite lane, and that it appar-

ently agrees with the total time for pass-ing.. This indicates that drivers allowthemselves very small margins of safety.Perhaps it should be added that due tothe method of analysis, the time I foundto be required for passing was the mini-mum average for all drivers and shouldbe compared with the $0 percentile valuefound by Forbes and Matson.

overtaking and passing requirements as...

Documents