' l “‘1 "1 h. ufi " M 2": '3 t a: Ma“ ! a5“! FL. 65“. Emmy AN”. t... MW...“ #4. . . .. mum .ng Jr «R an... ”mm {M . .9! a.“ ... Cava- m...‘.‘.§ fit. “to a... n .2 ”n. flu SP0. ma?) ... «\KRC . 1‘ o m ”m” tun! ... N... R... ya}: a 33 1.5... a... «an: at... :1. an rt... fin: s u .l U U I ..C” .1'. ‘1... .3, Ln. ~00! pv- WvU 5.1.... .293. AN... 19.... 3‘ Info“ n 0 tr” KER S :3? far 43%?! ‘3 F -- o \ . ‘ v. ...a” v ; ~94. .mk H» L A. G t Cu. ”5...“ a.” A. 0" ' .. i L r: h . no. .2...“ . 01—.“ Nut”: #II. 8 5‘0? I .45 n1 ‘2 U K H,E:E;:2;13.32:, \‘HEEW Michigan State University ABSTRACT EVALUATION OF TWO-LANE RAMP MERGES ON URBAN FREEWAYS BY Grovenor N. Grimes A study of two-lane ramp merges on urban freeways was made in an effort to analyze the design and operational characteristics of this type of merging situation. Sixteen millimeter color movie films were taken at four locations in the Detroit area. Vehicle path distri- butions were taken off the film by means of observing vehicles on a projected grid layout. Given projected traffic volumes, a method of ana- lyzing the capacity or traffic carrying ability of two-lane ramp merges was deve10ped based on the distribution of approaching ramp and mainline traffic. Since only one lane is added to the freeway beyond a two-lane merge, it is necessary to eliminate one of the ramp lanes. This study indicates that when the inside lane is dropped, traffic makes more efficient use of the total merge area. Further research is needed, however, to verify this conclusion. Grovenor N. Grimes ‘ Two of the two-lane ramp merges studied have been signed down to one lane because of Operational problems created by the design of these merges. The signing proved to be 90% effective; however, the capacity of the merge is very limited. EVALUATION OF TWO-LANE RAMP MERGES ON URBAN FREEWAYS BY Grovenor Ni Grimes A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Civil Engineering 1971 I, n Ll / a X) I", I]: ACKNOWLEDGMENTS The Michigan Department of State Highways supplied the films used as the basis for this thesis. I wish to express appreciation to Dr. Gail Blomquist, my advisor, for his assistance in the completion of this thesis. ii ACKNOWL LIST OF LIST OF LIST OF LIST OF Chapter I. II. III. IV. _BIBLIOG TABLE OF CONTENTS EDGMENTS . . . . . . . . TABLES . . . . . . . . FIGURES . . . . . . . . FORMS. . . . . . . . . GRAPHS . . . . . . . . INTRODUCTION. . . . . . . REVIEW OF LITERATURE . . . . METHOD OF STUDY. . . . . . Data Source . . . . . . Locations. . . . . . . Film Analysis . . . . . ANALYSIS OF RESULTS . . . . Capacity Analysis Merge Design. . Signing . . . . . . . CONCLUSIONS AND RECOMMENDATIONS Conclusions . . . . . . Recommendations. . . . . MP HY O O O O O O O O 0 iii Page ii iv vi vii 14 23 23 33 42 46 46 48 51 Table l. 2. 3. US-lO Merge: M-39 Merge: 1-75 Merge: I-94 Merge: LIST OF TABLES Time and Volume Data. Time and Volume Data Time and Volume Data Time and Volume Data iv Page 19 20 21 22 LIST OF FIGURES Figure Page 1. General Area Map . . . . . . . . . . 9 2. US-lO Merge: Geometric Data . . . . . . 10 3. M-39 Merge: Geometric Data . . . . . . ll 4. 1-75 Merge: Geometric Data . . . . . . 12 5. 1—94 Merge: Geometric Data . . . . . . l3 6. Typical Grid Layout . . . . . . . . . 15 7. US-lO Merge: Percentage Distribution of Peak Hour Traffic and Nomographs. . . . 28 8. M-39 Merge: Percentage Distribution of Peak Hour Traffic and Nomographs. . . . 29 9. US-lO Merge and M-39 Merge: Percentage Distribution of Off-Peak Hour Traffic . . 30 10. 1-75 Merge: Percentage Distribution of Off-Peak Hour and Peak Hour Traffic. . . 43 11. I-94 Merge: Percentage Distribution of Off-Peak Hour and Peak Hour Traffic. . . 44 LIST OF FORMS Form Page 1. RecOrd of Vehicle Paths from Film Analysis . . . . . . . . . . . . l6 2. Record of All Vehicle Paths per Film. . . . 17 vi Graph 1. US-lO Merge: Traffic in US-lO Merge: Traffic in US-lO Merge: Traffic in M-39 Merge: Traffic in M-39 Merge: Traffic in M-39 Merge: Traffic in LIST OF GRAPHS Percentage Distribution of Lane A . . . . . . Percentage Distribution of Lane B . . . . . . Percentage Distribution of Lane 1 O O O O O 0 Percentage Distribution of Lane A . .- . . . . Percentage Distribution of Lane B . . . . . . Percentage Distribution of Lane 1 O O O O O 0 vii Page 34 35 36 39 40 41 CHAPTER I INTRODUCTION The purpose of this thesis is to evaluate the capacity, design and operation of two-lane ramp merges on urban freeways. In the Detroit area there are in exis- tence a number of high-speed directional freeway-to-freeway interchanges which contain one or more two-lane entrance ramps. Very high volumes can be carried on two-lane ramps. To handle these high volumes, a freeway lane is generally added to the freeway laneage at the end of the merge. This thesis will include only those two-lane ramp merges where a lane is added to the freeway beyond the merge. " Since very little research has been carried out in this area of freeway operations, detailed analysis of capacity, geometric design and operation is very difficult. The following problems will be analyzed and dis- cussed in detail: 1. In the early design stages of urban freeways it is necessary to evaluate the ability of the —freeway to carry projected traffic volumes at a specified level of service. Any merge point is a potential bottleneck and must therefore be analyzed to determine if the projected total 1 merging traffic is sufficient to cause severe congestion and possible stoppage of the freeway. Given projected traffic volumes, a method is therefore needed to determine the capacity or traffic carrying ability of two—lane ramp merges. 2. Since only one lane is added to the freeway be- yond a two-lane merge, it is necessary to elim- inate one of the ramp lanes. There has been considerable controversy for a number of years as to whether the outside or inside lane should be eliminated. An attempt will be made to pro- vide a rationale for determining the proper course of action. - 3. Two of the two-lane ramps studied have been signed down to one lane even though a two-lane merge was constructed. Operational problems were experienced after construction due to the approach ramp and freeway geometrics. This section will evaluate the effectiveness of the signing and the effect on the capacity poten- tial of the merge. In order to analyze the operation of two-lane ramp merges, 16mm color movie films were taken at four locations in the Detroit area. Film analysis of vehicle paths pro— vided the data needed to accomplish the goals of this thesis. CHAPTER II REVIEW OF LITERATURE A complete review of available studies indicates that very little extensive research has been carried out on the subject of two-lane ramp merges. In 1959 C. J. Keese, C. Pinnell, and W. R. McCasland of the Texas Transportation Institute (6) pre— sented a preliminary report, at the 38th Annual Meeting of the Highway Research Board, on freeway entrance ramp operations which included one two-lane merge. Vehicle paths for 713 vehicles were analyzed indicating that minor use was made of the ramp as a two-lane facility. No other conclusions were documented. Joseph W. Hess (4) in 1963 published findings of a nationwide study sponsored by the Bureau of Public Roads and the Highway Research Board concerning "Capacities and Characteristics of Ramp Connections." Two-lane ramp opera- tions (both on and off types) comprised 42 of the 219 separate studies submitted. However, the two-lane ramp operations varied so widely in both geometries and traffic characteristics that no capacity formulas could be deter- mined. Mr. Hess did conclude that an extra downstream freeway lane should be added beyond the merge. The 1965 Highway Capacity Manual (3) published by the Highway Research Board contains procedures for ana- lyzing all types of highway capacity problems. In the chapter on ramps, two-lane ramp merges are discussed. Four different cases are presented to analyZe four differ- ent designs. Case I--This design requires the addition of a free- way lane and provides the outside ramp lane with direct entry into the added freeway lane. The inside ramp lane must merge into lane 1 of the freeway or into the outside ramp lane. Research results re- garding performance are not yet available, estimates therefore are necessary. The Capacity Manual suggests that for Case I the outside lane carry the bulk of the traffic up to its capacity. The remainder of the traffic will be in the inside ramp lane and will enter lane 1 of the freeway as if it were a single-lane entrance ramp. Case II--This design also requires the addition of a freeway lane. In this case, however, the inside ramp lane is led directly into the added freeway lane the outside ramp lane is expected to merge with the inside ramp lane. Again, research results are unavailable. A general computational method for this type of design cannot be suggested, inasmuch as marking practices can affect the paths followed by ramp drivers. Cases III and IV do not require an added freeway lane and will, therefore, not be discussed. Research is also lack- ing for both of these cases. "A Policy on Geometric Design of Rural Highways" (2) published by The American Association of State Highway Officials virtually ignores two-lane ramp merges. They do point out some general capacity limitations for the ramp proper which may or may not be critical, since the merging point with the freeway will generally control the amount of traffic which can be handled on the ramp. "A Policy on Arterial Highways in Urban Areas" (1), also published by the American Association of State High- way Officials, states: In conjunction with entrances bringing 2 lanes of traffic into a highway, the highway beyond the ramp entrance should be at least one lane wider than the highway approaching the entrance. In a 1968 issue of Traffic Engineering, two-lane entrance ramps are discussed by Ronald C. Pfefer (8) based on conclusions by the S-F Committee of the Institute of Traffic Engineers. While a considerable amount of data was obtained by the committee, Mr. Pfefer states: It should be emphasized, however, that the conclusions are not the result of extensive research and should not be considered as such. The need for basic data to establish Specific design criteria is evident. While capacity is the common warrant for a two-lane ramp, there is a lack of available data on which to base capacity analysis. The current use of three merging lane configurations .further complicates the problem. In general, however, capacity must be checked at four points to assure adequate design; on the ramp proper, at the diverge, at the merge and on the freeway beyond the merge. Mr. Pfefer's article includes a simplified procedure, de- ve10ped by J. E. Lelsch, for analyzing inner-lane merge designs based on the 1965 Highway Capacity Manual. It assumes, as previously discussed in Case I of the Highway Capacity Manual, that the outside ramp lane operates directly into the added freeway lane and carries the bulk of the ramp traffic. Using the Capacity Manual procedure, the inside ramp lane is analyzed as a normal single-lane ramp merge. The two values are added to compute the total ramp volume. The article goes on to discuss geometric design and Operational considerations when determining whether to use the inner lane merge or outer lane merge. Mr. Pfefer further states: Highway agencies indicate varying design standards for the merge. In most instances, however, a direct taper of 50:1 or 600 ft. was utilized. Standard ramp width was 24 ft. The committee concluded that the ramp lane carrying the greater share of the traffic should be carried directly into the added freeway lane. This is especially true if the ramp is short and traffic has very little time to maneuver and change lanes. CHAPTER III METHOD OF STUDY When considering a method of study to evaluate two- lane merges, it is immediately evident that a complex Operation is taking place. In comparison with a single- lane merge, where the ramp vehicle is faced with a simple task of finding a large enough gap in lane 1 to fit into, drivers in the ramp lanes of two-lane merges have the Option of either proceeding straight ahead into the added lane or merging into freeway traffic. These movements are coupled with that traffic in the outside lane of the freeway which desires to shift into the added lane to gain more freedom of movement or to exit the freeway. Data Source In order to document the various movements taking place in the merge, it was felt that a permanent record was needed to insure a complete and accurate data source. It was therefore decided to use 16mm color movie films taken at 8 frames per second. At each of the locations, fOur reels of film totaling approximately one-half hour in time were taken during the peak hour and four reels during an off peak hour. All filming was done during the summer \ in good weather, on a normal working day. 7 The camera was set up on the top platform of a maintenance truck which was parked at a location providing good visibility of the merging area. It was generally found that setting up well back in the gore area of a merge and shooting downstream provided the most distortion-free films which were the easiest to analyze. Before filming, each merge was marked with orange cones and orange paint F1 to provide a reference point every 100 feet. Locations Figure l is a general area map Of the Detroit Metro- 2 g) politan Area and shows the four locations which were uti- lized in this study. Although there are more than four two-lane merges in the Detroit area, the ones chosen were constructed in recent years and therefore are considered to be representative of newer design practices. Further- more, economic considerations prevented filming any more than four locations. The locations and pertinent data are shown on Figures 2 through 5. For convenience, they will be referred to in the following manner: ' 1. US-lO merge 2. M-39 merge 3. I-75 merge 4. I-94 merge It should be noted that throughout this thesis lane A refers to the outside ramp lane, lane B the ramp lane closest to the freeway, and lane 1 is the outside freeway lane. 7‘9 (.‘O f OAKLAND ‘ MACOMB I. IIILE IO :3 Vacuum mo '1 “CNN“? \ G'LKIT S 2 ' *‘TM—‘w cannon-I 8 l4 MILE l “Ill“ 0 mm“ ”sun can ' *__-. 9 MIL: 7 MILE [MGRII 3 E F; a O 3 U'l -I rtnnELL (mum. '0: I" ., _ 0MPM'IR “WINGS M GIEINIILD W K. "Frin+ Lugccxu M ‘ 13 “UL! .. . - . WHATS”... . FERNDILP 0 ~——~—®‘ L\\}' uszLIIE. ‘now 7 MILE I a ‘1' 36 \ . - n!” "000, "M moms ° 3 HAZEL PARK‘ -14 ‘1‘! sun null no ~ JON" I \4v, "7. v K NICHOLS ummu M .'"‘ a WA YNE {ESSEX 8 Windmill Point meme“ ‘3’ ’Rbfii'lfit" 0mm 5 us. . UN. 10 .muma afiuuwsowo BE-W “mane: CHImDII.N MMDun // \/ .cwmm Osmulmmo Emwuumc30p Ou muscumwn ow amoz um chaum>uso nswm uswmcme mmoz um chaum>uso mm3owum men H+ omoz um compo mmsmmum Ame H+ mmoz um mpmuo seam mu on N moswwuw>soo mo mec< kpswm .NH supfiz mama hwzmmum m .N~ cubes msmg seam cowumooq muoEmo VI Magi ALI. . mm 6:3 . III i. F ucoq l w m 93.. L .oov .wumm UHhUQEOQU 9 muwm 0 .' 35pm / Hausa: amnqu.m mxSOHm .mcom Oswplmwo Ecouums30p ou mosmumwm ON mmoz um wusum>pso mama H usmwsmH owoz um mucum>usu masooum Nmm.al mmoz um opmuo zmaomum z NNN.HI omoz um opmuu mama om.H mosowum>coo mo oawc< .NH rupws mama hwammum .NL rue“: mama mama < 33H l II} '1 ‘m econ 1" wucom .lll mecoHI r?! "LA 1L 4/ .oom mwm // COHDOOOA mHmEmo .mumn afiuuosooo 12 "mwum: mkuHII.s mmpuHm .mwuwa w:u mo mocm>vm ca < mama Ousw Ofiwmmuu m mama umwsm ou Dames ma paw whom oxu mo mosw>pm :H .oom hamumawxoud lam Oammmuu Osmu mo ummH orb On each msu so pmumooa OH swam mHLH .ccmm .coqa Om usmmsme xcc.HI Noe.01 om.m .NH .- OEmuIcO IIIIIIII|||I|I|III iIII‘IIIIIIIIIII “wuoz camulmmo Emmpumcsop ou mocmumwn EmmuumsBOp ou mosmumwn omoz um musum>uso deem mmoz um musum>uso hmamoum omoz um mpmuu hmaooum mmoz um womuu seam Losmwum>soo mo mawc< zupHB mama hmsmmum zupfiz mung neon cowumooq mama—mu \ I\l\\\\\\\ .IMHMMHHHHIIIIIIIIIII 4 econ F OCOD “(6:04 m ocnj 13 .mumo unnumsooo "mayo: smIHII.m mmpOHu .ouow OSD acarowmu whommn 4 Mesa can“ cam Immuu m mama Omazm ou usage ma swam were .muow Osu mo mosm> 1pm sfi .ocm %kum5«xoummm cam Immuu damn mo ummH ocu ou dawn Ono so pwumooa OH swam OHLH ”wuoz «Dam .mecm Osmulwmo Emouuwssop ou oosmumwm owoz um musum>u=o dawn ¢ ucmwsOH Omoz um musum>uso mmBomum woe m OJ 0 u m NOH.H+ wmoz um upmuo kmzmmum w m Ncm.H+ mmoz um mpmuo mama v.1 ON mosmwnw>soo mo oawc< m .NH rue“: mama smammum .NH sue“; mama asmm . OOHOOOOA mumSmu 1llIlIlIIIIIIIIIIIIIIIIIAVMIWII 14 Film Analysis The use of films as the data source provides a permanent record which can be used over and over. However, ,transferring the film data into usable information was both difficult and time consuming. A Perceptoscope 16mm film projector was used to analyze the film. This projector enables the Operator to run the film forward or backward at any Speed. It can also be stOpped on any frame desired. Film analysis was accomplished by projecting the picture on a screen at a fixed magnification. A grid lay- out Of the merge similar to the one shown on Figure 6 was then drawn on the screen, matching the 100 ft. field refer- ence marks. Joint lines were also shown on the grid. Form 1 was used by the observer to record the path of each vehicle as it proceeded through the merge. If a vehicle enters the merge in lane A, the Observer records ,the vehicle‘s entering lane and each point at which the left rear wheel, if merging left (or the right rear wheel if merging right), crosses a lane line. To illustrate this process a theoretical car path is shown on Figure 6. 1 This path is recorded on Form 1. The car enters the merge in lane A, the left rear wheel crosses the first lane line at point 3 and the second lane line at point 17. Form 2 was used to total the vehicle paths recorded on Form 1. The segments shown on Form 2 are the same as 15 .uOOhma capo HOOHOAHII.© mmDon 16 Appr. Crossing Appr. Crossing Appr. Crossing Lane Points Lane Points Lane Points fl 3 / 7 } - II I, ~ .- 4? ~—---—-u—— r— i ------ I --- I ...—-..-..” -4 r...» ...-1%. ...... l -.__ __._. _. l4}. _ )- «-~- -~—- — F" ~ I— ——- ‘1?’* I-- - —-~<)—-—-— .___... ( ...... F r—_____....-V...I . ...... 0-— 1-..- .I .- I .. I .. ——-——~ -———-—1;—— —- -——— -~~—-—-——-—q———---~~—-—»-- ~-0'~— .... ~-~-- - ~— - —-~-——< -—- - - —- JL « _ _H _ h_ ”1' - ‘._.__ I” ...... i” -_ 1 _ _ 1 - _- 1 - -—-- - *-—— —--j>»--— -— 1—~- -- »-Ir- - i F.__._.._ -... J 1. _ l”. -... .1r--___.J-..-._ .. FORM 1.--Record of Vehicle Paths from Film Analysis. Lane A c " -.,‘ 7 Lane ‘chmCHL I l 1 Segment 2 17 :Segment 1 3 Segment 4 """1? Segment 5 Segment 6 rOTAL 1 Lane Line Lane B Lane Line Lane Line Lane 1 Lane Line Lane Line FORM 2.—-Record of A11 Vehicle Paths Der Film. 18 those shown on Figure 6. Form 2 was completed for each film. Again, using the same theoretical vehicle path, the crossing point at 3 is in segment 2. A mark is therefore recorded under segment 2 as shown. The crossing point at 17 is in segment 4. A mark is recorded under segment 4. The two marks are then connected with a line to denote the path of that particular vehicle. Each vehicle is recorded 'l'f.‘ l.'. i' t" in this manner. Having determined the number of vehicles in each segment, the percentage distributions were developed as presented in the chapter on Analysis of Results. E} A total of 7,742 vehicles were Observed in this manner. Tables 1 through 4 indicate the actual number Of vehicles analyzed at each location. Since the films vary in length, it was necessary to expand the observed volumes to vehicles per hour as shown. The expanded volumes pro- vide a direct comparison between the peak and Off-peak periods. Vehicle speeds and gap acceptance were not analyzed in this study. However, since a permanent film record is in existence, further insight into the Operation of the merges could be obtained as part Of a future study. 19 momumhm own mam 0mm «mm m own mmaa cam nmm we mm om ma mv.m menu mans v H mom own mmm 3m moa moa 3. ma oa.m omum mass m m. was mmm oav mmw moa baa am am am.b maum main m M ham men amm men we aoa Na ma vm.n oonm mans a x anma puma mum hum mmmum>a mmna mvea mam ovoa ova mma oma voa mo.m maum mmlm v m maom mo: mom mom Mum amm mma moa. 34w Houm mmnm m M“. mama omma omm omm mmm mam «Na vma oo.m mvuw mmlm N m mama mama awn mmn mam mom ooa mm mm.h omuv mmum a N H m ¢ N H m fi ASHSV Ava mama mama mama mama mama mama mama mama maaa mafia # amw‘meaao> ompammwm Edam a0 mfiaao> Eaam mumum mama ammm .mumn meaao> cam mafia ”mmumz oaImDII.a mamas mew mew boa mew mmmum>a 20 O T: 444 oma qua mam em ow ma km oo.m manoa man“ a H mma mow Goa Nmm mm mm 4a ma am.e omuoa mane m m mam mow om 6mm am mm m as mo.m manoa maue m m ham mmm maa mav mm mm ma am mm.k couoa mane a x mam nmo amn wwoa mmmnm>¢ oms mam 0mm mam am we om mma aa.m omuk mass 4 m m; mm... mg 2.: S: a: a: 3a 35 mm; 3A m m Ham amm as“ mam moa mm mm mma mm.h omna maus m m mmm was man amma mma aoa moa sea mo.m monk wank a m a m m m a m a Aaazv “say mcmq mcmq mgmq mCMH mqmq mcmg mama 05mg mEflE GEHB # am> mEaao> Ombammxm Saab a0 mfiaao> Eaam mumum mama ammm .muma 055H0> Cfim QEHB "magma mmuzau.m mamas 21 aom mam new mom mmoa mmmamza m. mmm omn «ma mmm aoma om mm mm am mma wa.m mmuaa mmlm v H mam mmm mmm Nam bwm am ma om mm mma mmum omuaa mmnm m m. «mm mom «av mna mam ma on am an hma mmum mouaa mmum N M mmm mmm com hma mmoa mm am om ma maa «mum omnoa mmlm a 1 mama moma maoa oma mama mmmum>a mama moma mam amm aoma aha mom maa av mam mm.m ooum mmlm v m mNma mooa mmoa ova mmha mom mmm mma om 0mm mm.m mans. mmum m m. vama mona moma moa amma ama mmm mma ma mam w~.m omub mmlm m m onaa aoma mvaa mm vmma oma mam nma ma mam am.w mauu mmlm a m N a m a m m a a Eda ES mama mama mama mama mama mama mama mama mama mama msaa man # am> meaao> pmaammwm Edam ao maaao> Eaam uumum mumo ammm .muma maaao> mam msaa "mmumz mhIaII.m mamas 22 omm owe mam ma ama mmmum>4 o mam mmm mmm mm mow am mm mm m mm am.m oouaa mmum a m mha mma mmm Ha avg am am av m om aa.w mauoa mans m m mmm saw mam ma mna mm mm om m we am.m omuoa mmam m m «ma «ma amm mm aha em mm mm m we mm.m mauoa mmum a m now 5mm awn mm nmm momam>a mow mam mam so mam em mm mm m ma mv.m Nana mmum a m pom mum mom am mam om mm mm om ma.m «mum mmum m m Ham Nam mam as mam ma Ne av as av.m amum mmum m m mam mam 5mm on mmm me an am mm mn.h houm mmum a m N a m a m m a m a Aaazv 124C mamHI mama mama mama mama mama mama mama mama mama mEaB mEaB * am> mEaaO> Ummammxm Eaam a0 mEaao> Eaam uumum mama ammm .muma maaao> mam mEaB "mmumz vaHII.v mamaa CHAPTER IV ANALYSIS OF RESULTS Capacity Analysis The subject of capacity analysis is a complex one. When discussing capacity, confusion exists even among traf- fic engineers concerning the meaning Of the various terms used. This thesis will use the 1965 Highway Capacity Manual (3) as the basis for defining the terms used in this chapter. The 1965 Highway Capacity Manual defines capacity as "the maximum number of vehicles which have a reasonable expectation Of passing over a given section of freeway during a given time period under prevailing roadway and traffic conditions." As a rule of thumb, the capacity of a freeway lane is 2,000 vehicles per hour. Naturally, a freeway does not Operate at capacity at all times. The Capacity Manual therefore deveIOps the concept of level of service, which is simply a measure of various operating conditions ranging from very light traffic (Level of Service A) to stop-and-go traffic (Level of Service E). The Capacity Manual describes the various levels of service in the following general terms. 23 24 Level of Service A describes a condition of free flow, with low volumes and high speeds. Traffic density is low, with speeds controlled by driver desires, speed limits, and physical roadway condi- tions. There is little or no restriction in maneu- verability due to the presence of other vehicles, and drivers can maintain their desired speeds with little or no delay. Level Of Service B is in the zone of stable flow, with Operating speeds beginning to be restricted somewhat by traffic conditions. Drivers still have reasonable freedom to select their speed and lane ‘ Of Operation. Reductions in speed are not unrea- .I sonable, with a low probability of traffic flow 1 being restricted. The lower limit (lowest speed, : highest volume) of this level of service has been 1 associated with service volumes used in the design L of rural highways. i Level of Service C is still in the zone of stable flow, but speeds and maneuverability are more closely controlled by the higher volumes. Most of the drivers are restricted in their freedom to select their own speed, change lanes, or pass. A relatively satisfactory operating speed is still obtained, with service volumes perhaps suitable for urban design practice. Level of Service D approaches unstable flow, with tolerable Operating speeds being maintained though considerably affected by changes in Operating con- ditions. Fluctuations in volume and temporary restrictions to flow may cause substantial drops in Operating Speeds. Drivers have little freedom to maneuver, and comfort and convenience are low, but conditions can be tolerated for short periods of time. Level Of Service E cannot be described by Speed alone, but represents Operations at even lower operating speeds than in Level D, with volumes at or near the capacity of the highway. At capac- ity, speeds are typically, but not always, in the neighborhood of 30 mph. Flow is unstable, and there may be stoppages of momentary duration. 25 For purposes of this discussion, the maximum volumes which can be handled at each level of service are as follows: Level of Service Vehicles per hour per lane (5% trucks) A 1,000 B 1,200 C 1,400 D 1,600 B 1,800 Level of Service D is generally used when designing free- lways for the Detroit area. The ability of a freeway to handle the design level of service is dependent on the merge and diverge points. For example, at the merge of a single-lane ramp and lane 1 of the freeway, if the total volume of the two exceeds the design volume, then the potential exists for a bottleneck. Either the ramp traffic will back up, or freeway traffic will have to shift out of lane 1 assuming there is room in adjacent lanes. For Level of Service D, at the merge of a single-lane ramp and lane 1, 1800 Vph is considered to be the capacity at that point. Any volume exceeding 1800 vph will cause the merge to break down. Moskowitz and Neman (7) have described the capacity of a merge in the following manner. Merging Operation will be smooth as long as total ramp and adjacent lane rate-of-flow does not exceed 1800 vph provided that the entrance ramp terminal is long enough and has a gradual taper. Maximum combined flow-rates for a merge of a parti- cular ramp and adjacent freeway lane have been Ob- served as high as 2000 and 2200 Vph. However, it 26 is not recommended that this figure be anticipated in design procedures, since there are certain condi— tions of geometric design and traffic characteristics (which are difficult to predict or evaluate) that can prevent its attainment. 1800 vph is a depend- able figure and can be counted on under all circum- stances, with normal truck percentages and grades of less than 3%. Therefore, 1800 vph is the key to analyzing any merge Operation and will also be the basis for analyzing the capacity of two-lane merges. Two-lane ramp merges involve the merging and crossing of traffic in lanes A, B and 1. For purposes of analysis the lanes A, B and l were divided into 100 foot segments. At any particular segment along the merge, a portion or percentage of the traffic from each approach lane will be in that segment. If this traffic totals more than 1800 Vph then that segment will experience congestion and possible stop-and-go operation, which in turn effects the rest of the merge. It was therefore necessary to determine how the ramp and lane 1 traffic distributed itself when passing through the merge area. This was accomplished by compiling by percentages the vehicle paths taken Off the films. For each 100 foot segment the percentage distribution of traffic originating from lanes A, B and l was compiled. It was determined that the peak hour traffic should be used as the basis for the capacity analysis, as this would be representative of typical everyday urban driving conditions. The results of the film analysis of peak hour traf- fic for the US-lO merge and the M—39 merge are illustrated 27 in Figures 7 and 8 respectively. Figure 9 shows the Off- peak distribution for the two merges. The design of the US-lO merge provides for the elimination of lane B and is referred to as an inside merge. The M-39 merge provides for the eliminatiOn of lane A and is referred to as an outside merge. Both will be analyzed for comparison of their traffic carrying ability. In developing this capacity analysis method, it !l was felt that one segment in each lane would exceed 1800 I vph before the other segments in that lane, regardless E of the distribution of traffic entering the merge. It was therefore necessary to check all the lane-segments for all possible combinations of entering volumes to determine the critical segments. This critical segment could then be used for determining the capacity of each lane. The criti- cal segments found in the analysis of M-39 and US—lO are indicated on Figures 7 and 8. It is now necessary to check only three segments for the US-lO merge and two segments for the M-39 merge to determine whether 1800 Vph will be exceeded in the merge area. For the US-10 merge, the nomographs shown on Figure 7 provide a simple method of determining whether the merge will exceed 1800 vph. The nomographs were developed using the percentages found at the critical segment of each lane, based on a method found in Hoelscher and Springer's "Engineering Drawing and Geometry." For example, the nomograph for the lane A merge volume is based on the percentages found in the critical Critical Segment lUU' ‘l)' I /, fl ‘7' , A“. I t‘ 500 . ‘i‘lu' )llt)’ \ hill]. ’0‘)! ewx__ __. .o . e l T? VJ I I“ “:I' . _ h .7» _- _ 7 l.._l__~ V » \ZQQ I \\/:vf—\: D V V0 3 -— '.' >1 , A") l I \\l 1 _> _ an _. _ ,1 A... -.. / w— ---- /‘~‘ ' ——_—_ ,7’ \ ‘II ‘ ~ \' \‘.;;Ji____—va ~’——-a‘"1‘-‘ ‘-———" <§ 7/. 7‘1 \;y 1 r'; U. ‘ fl ‘1 “/l f \, \V \lfll) I.._J ‘ _ - __,_fl_’-L-""—' “NU-‘Iflfiu—w— ‘ \/7/V, \\;_/;/{/l} l . ‘ I I. ) l‘ / ’ ll. NJ ‘ l..‘\\l A F, ”"f 1. 1 if , “3Y1 ~\r- ~~ . ..., ’ ""’ “'7 \ H V» \'\/(~9 \«f '\\' ' A \.,./fL—V—J_/_____,._._,_._. __A _ '_ , “*4 /_,__,_.._.._ ~- — ,,_i______—~ _-__. 4,- - e _ lle-‘ll 'I' I . ,i._ 'd" '“ \ AX ‘ I ‘ I_ ‘9“ '-_ _ »'_ ‘_ .___<.,, '- ' __”""’_~F _7+ “ ‘ "/f’_"-" y.\\ ‘ \/\\/‘ ___.,,__—~————-—->—‘\—‘2— M'~—> ‘7‘»- 15V”. :‘\ ‘fl/ _ -' “l J, -- ’ " '_ _"' (:I’lLiL‘Jl / ,- . - " 1// Segment CI' ll lL'.|l Segment V 15 Rump Lune IS Appronch Volume V v.p.h. l Freeway Lane l 1F a 100 F Approach Volume ' v.p.h. T7800 Solution Solution l’rocedure: HI- r y, e Vo ] um e V V + V To determine the lzine A merye volume ' ’ I ‘ ~ A + ‘ \A \B ”900 7» V4)?!“ 1 l. Draw .1 line from VA to Vll' Romp l...l11t' A \, R.imp Lune B . - .. . - -. . - - . > , ‘ . A 600 Inteiaecting tine turning line. Appladtl‘. V\olume 1; Approach Volume -. , ‘ 3. Draw .1 line from Step 1 inter— V'E'H' \'-l"h' CAUU Section of the turning line to Rump Lane is ‘ . V.:CO l Vt App roach Volume 1900 ~ - T \'.p.l1. ‘ RUIULVIUH 3. The intersection point of the FCC) ”1"17‘6“ Volume lIJOO .. ‘L (CHE/K) line from step I with the T i , "A + I.“ V 41400 solution line is the total v.p.1. - . , A meIgL volume. 1800 lbUU‘ Rump Lnne A ,~ 1600 4)- T 4* Approach Volume (Lil-1K- . v.1).h. To determine the lane B merge volume . 4. 1[000 l l. Draw ;I line from VA to VB T1800 1200 . , 1200 . .w.-- , -.. 1 ‘ i'~ _ . .. ”50b I lllLLlattLlllt, tnL so ution inc l'VUO 4+ . leO 2. 'l'he intersection point ol the 100 a lJUO 4 line from step 1 with the .. 4 . . . . r . solution line 15 the total 7400 .IJOO ,r v C1 U merge volume. Sullll loll \A 1 00 g H‘ | (J ’) Merge Volume R.1mp l H“. A 4 4’ I» 1‘00 j 4 1000 T , , , “ ~ . ’ \A : \“hl \1 Al‘l"“»’-'“'11 VOIUHIU 1200 1000 w. 000 ll 1 To determine the lane 1 merge volume LOO j_ '1’ ' lqOO V.p.h. 'l " C 4_ l~+UO l. RL-pe.It the procedure for deter- L \' '”‘ . . . ‘ l , a llllnln): the l.lIlt' A merge volume. 1800 1.00 l :00 I J y Freeway Lane 1 1' ‘ ‘0 ,3! . Lug ' ANII'o.Ieh Volume 4 lQUU bOO ‘ 17L'U 800 v.p.h. ,‘ ‘ 0 ‘— ‘ 1000 1500 I ”1300 +1000 q , + 1. ,, TOOO 180C . moo 1 4. 400 eOo ., - l , 0 J10 000 f / _ 100 OD r 1000 l‘400 ‘ eoo ‘ UUU i U ” ‘ l4OO 1L 1’) L4- 800 L .... 800 .00 ‘ lob VOO + «II— 800 800 1200 . . 1000 0 2V0 . 500 + : l 1 800 J O '3 ~ oc‘ I“ OOO .. i» 600 4 O 600 . 1000 “ I COO ‘ 0 ..bOO I 0 -(I- t-E ,, 5 e00 ” 400 C t 4* 2 ‘ 4 400 g 00 “:00 40'0 4> L H «y 600 - - 400 O 4 OO )- ~ 4_ 400 20L) 0‘ 400 l ZOO 200 . 200 i . 200 200 0 . O 200 ..200 l o O J' -t O O .1- LANE A MERGE VOLUME LANE E MERGE VOLUME LANE l MERGE VOLUME FIGURE 7.——US—10 Merge: Percentage Distribution of Peak Hour Traffic and Nomographs. 29 Critical Segment .00' 500' 600' 700. '“VOE E VIE' ‘ V®e at. i &4 VJ ‘ \Critiml Segment I3 79 LANE l —-> v .s B 4—— 95121 LANE 8*: PERCENTAGE DISTRIBUTION OF PEAK HOUR TRAFFIC V A Ramp Lane A Approach Volume l Vop.ll. { 1500 Solution Procedure: l T V To determine the lane B merge volume Solution B Draw a line from VA to VB intersecting Merge Volume Ramp Lane B the turning line. VA + VB + V1 Approach Volume Draw a lino-fromistep 1 intersection lFMDO v h V " n.h. V of the turning line to V1. v ‘F '9. . l . l The intersection point of the line from "F‘2400 Freeway Lane 1 "F Freeway Lane 1 step 2 with the solution line is the Approach :olume T lBOC) Approach Kolume total merge volume. V Vopu- ~ Vol). 0 B .. ' , . [‘0 determine the lane 1 merge volume 1"400 _ Ramp Lane B - 1800 ‘ 1800 1. Repeat steps 1 through 3. ‘ Approach Volume 7 Solution [7 __ v.p.h. V _; 1600 Merge Volume , 1 “ A V + V + V “r 1600 q— 600 Ramp Lane A A_v.p1.3h. l it 1600 lDOO Approach Volume 2400 1:2 5 w- a A o u t -‘ OO _. P lbOO 4'- 14C)O V p h .0. 1400 F -. 1500 i 1400 1800 T v 2200 .. 1400 1200 ~0- ldOO 1600 ‘L 2 A 1000 i .-1400‘" “1200 —— 090 12 1000 1400 ~ ‘ I 00 .J. 1000 -r , i m . 1800 a I. 1200 1200 o T . __&m -~ I a g 800 T i1000 1000 800 i w _ 2000 i “ “*1600 ” .5. I 1000 _. 600 I? g i 600 800 'E' _. 1400 5 I 400 , “ ..t BOO “ o 800 6 g i 800 600 _“_ I .0. 200 4- 4()0 OO "‘ a}. 1200 600 - 400 i ‘t 4- ,- .. 1000 i 600 l O i 200 . l 000 ZOO .. , i. 400 -, BOO 400 + - O ' '0 y i 400 4.200 + 400 i 600 ._ 400 A .Jl. O .1; 200 400 I. ‘5 "' 200 IL 200 J , l 0 J L O 0 I- " - 0 LANE B MERGE VOLUME LANE 1 MERGE VOLUME FIGURE 8.——M—39 Merge: Percentage Distribution of Peak Hour Traffic and Nomographs. 3O 100' 200' 300' 400' 500' 600' 700' LANE 1 LAN If B M-39 MERGE PERCENTAGE DISTRIBUTION OF OFF PEAK HOUR TRAFFIC 600' 500' 600' 700' » LANE 1 -—’ LANE A —' US-lO MERGE PERCENTAGE DISTRIBUTION OF OFF PEAK HOUR TRAFFIC FIGURE 9.——US—lO Merge and M-39 Merge: Percentage Distribution of Off—Peak Hour Traffic. 31 segment for lane A. The procedure for using the nomo- graphs is also shown on Figure 7. The following example using Figure 7 illustrates this procedure. Given: Total Ramp Volume-—l800 Vph. Freeway Volume-—3500 vph. Two-lane freeway (one direction) Determine if lane A, B or 1 will exceed 1800 vph in the merge. Step 1: Determine the distribution of approaching ramp and mainline traffic. Assume that lane A carries 80% and lane B 20% of the ramp traffic. At the beginning of the merge each lane will therefore carry the following volumes: Lane A = 1440 vph. Lane B = 360 Vph. Lane 1 = 1400 vph. (from Table 8.3, 1965 Highway Capacity Manual) Step 2: Using the nomographs on Figure 7, determine the merge volume for each lane. The results are as follows: Lane A = 1830 vph. Lane B = 440 vph. Lane 1 = 1440 vph. While lane A exceeds 1800 vph, lane B and lane 1 are well under capacity. The reader will note that the lane A and lane B approach volume percentages were assumed as 80% for lane A and 20% for lane B. Present research has not developed a method for predicting these percentages. To point out the critical nature of this assumption, the percentages will be reversed. Given: Ramp volume--l800 Vph. Freeway volume-~3500 Vph. Step 1: Step 2: 32 Determine the distribution of approaching ramp and mainline traffic. Assume 20% in lane A and 80% in lane B. At the approach to the merge each lane will therefore carry the following volumes: Lane A = 360 vph. Lane B = 1440 vph. Lane 1 = 1400 Vph. Using the nomoqraphs on Figure 7, deter- mine the merge volumes for each lane. The results are as follows: Lane A = 1730 vph. Lane B = 1430 Vph. Lane 1 = 2000 Vph. By reversing the percentages in lanes A and B, lane A is no longer over 1800 Vph, however, lane 1 is now well over 1300 Vph. The nomographs for the M-39 merge shown on Figure 8 were deve10ped in the same manner as the US-lO nomographs. The same example will again be used. Lane A does not have a nomograph as the percentages distribution in any of its segments never exceeds 100%. Given: Step 1: Step 2: Ramp volume--l800 Vph. Freeway volume--3500 vph. Determine the distribution of approaching ramp and mainline traffic. Assume 80% in lane A and 20% in lane B. At the begin- ning of the merge each lane will carry the following volumes: Lane A = 1440 vph. Lane B = 360 vph. Lane 1 = 1400 vph. Using the nomographs on Figure 8, determine the merge volumes for each lane. The re- sults are as follows: Lane A = not critical. Lane B = 1720 Vph. Lane 1 = 1460 Vph. 33 In this case none of the lanes exceeds 1800 Vph and the ramp will handle the traffic without congestion. However if the percentages are again reversed, lane 1 will carry 2200 vph which is well over capacity. It should be noted that the I—75 and the I-94 merges were not utilized in this portion of the study. Because of geometric characteristics which affected the operation of the merge, they were signed down to one lane and were therefore not representative of typical two-lane merges. "Merge Design One goal of this thesis is to provide a rationale for determining which ramp lane to eliminate when designing a two-lane ramp merge. In order to draw some conclusions on this matter, it is necessary to discuss in detail the characteristics of each type of merge. US-lO Merge Lane B, the inside lane, is eliminated at this merge. Graphical representation of the percentage distri- bution of merging traffic during the peak and off-peak hours are shown in Graphs l, 2 and 3. These curves were developed from the percentages shown on Figures 7, 8 and 9. The peak and off-peak curves are fairly consistent. It is evident that about 10% more traffic changes lanes during the off-peak than during the peak. This is to be expected since the gaps during the off-peak are large and allow the driver considerably more freedom of movement. The effect 34 ocoH .< one; :« cannons no nofiuanquuuwn owuucouuom Auoomv annex wcoa< oucauuan Home»: caumpuu.a :m4uu COG ccc CCB CCO ccn CCQ ccm GCN cog nuquIIJ xmomi \\. ELK [III A xmom xmmm mmCl) xmmmx XI 0364. used Illnulll. 0cmA "WMM \\‘L // mwnm // m v—WMOL\ [1 Lvllilll le/ / /I j xmwa\uw, 1i o~ cm on oq on or en om om Goa (z) UOInnqunsIa 35 .m ocog :« ofimuaua mo cowuanwuumfia eunucouuom AuGOHV annex unca< ooamunwn "ammo: OatmDII.N mmdmu CCCH Goa can can ccc can 00¢ 0cm CON can $ moo .rmmwmmuwiw xx” I \ 4. mm \_ A. soon x m \\ \ / I ti xmmm 4/\ HMO zoom HMO x l mama -. 5 1-341-, 33-3-; mama . .Iluulll MCMA xmom “rmu //. .-----ulf / . a. OH on on cc cm. or on ow cm ecu (x) “013“QT13910 36 coca .H one; cw oammaua mo coau=AwuumHn ommucoouom Auoohv omuoz waoa< oocgumwn "onto: canmann.m =m