Flexible Pavement Design Using The Asphalt Institute Method [The following pages, describing the Asphalt Institute Method, are excerpts from Traffic and Highway Engineering by Nicolas J. Garber and Lester A. Hoel, West Publishers, 1988]792 DESIGN OF FLEXIBLE PAVEMENTS. GENERAL PRINCIPLES OF FLEXIBLE PAVEMENT DESIGN In the design of flexible pavements, the pavement. structure is usually considered as a multilayeted elastic system, with the material in each layer characterized by certain physical properties that may include the modulus of elasticity, the resilient modulus, and the Poisson ratio. It is usually assumed that the subgrade layer is infinite in both the horizontal and vertical directions, whereas the other layers are finite in the vertical direction and infinite in the horizontal direction. The applica- tion of a wheel load causes, a stress distribution, which can be represented as shown in Figure 18.2. The maximum vertical stresses are compressive and occur directly under the wheel load. These decrease with increase in depth from the surface. The maximum horizontal stresses also occur directly under the wheel “load but can be either tensile or compressive as shown in Figure 18.2(c). When the load and pavement thickness are within certain ranges, horizontal compressive stresses will occur above the neutral axis, whereas horizontal tensile stresses will occur below the neutral axis. The temperature distribution within the pave- ment structure, as shown in Figure 18.2(d), will also have an effect on the magnitude of the stresses. The design of the pavement is therefore generally Figure 18.2 Typical Stress and Temperature Distributions in a Flexible Pavement Under a Wheel Load ‘Wheel load, W c Compression Tension =— = [esas ] ee ) (2) Pavement layers (b) Distribution of (c) Distribution of (d) Temperature Vertical stress” horizontal stress distribution Under eenterine Under centerline of whet oad of wheel load P= wheel pressure applied on pavement surface a= radius of circular area over which whee load is spread = compressive horizontal stress 1 tensile horizontal stress -DESIGN OF FLEXIBLE PAVEMENTS 793 Figure 18.3 Spread of Wheel Load Pressure Through Pavement Structure Wheel load, W or onan TOsEALE (a) Stress distribution within different (b) General form components of the pavement of stress reduction Source: Reproduced from Thickness Design—Asphalt Pavements for Highways and Streets, Manual Series No. 1, The Asphalt Institute, College Park, Md., September 1981. based on strain criteria that limit both the horizontal and vertical strains below those that will cause excessive cracking and excessive permanent deformation. These criteria are considered in terms of repeated load applications since it is known that the accumulated repetitions of the traffic loads are of significant, importance to the development of cracks and permanent deformation of the pavement. Figure 18.4 Schematic of Tensile and Compressive Stresses in Pavement Structure Wheel load; W » or Raw pavenent wee o—COlNERESSION — —om TENSION Source: Reproduced from Thickitess Design—Asphalt Pavemiénts for Highways and Streets, Manual Series No. 1, The Asphalt Institute, College Park, Md., September 1981794 DESIGN OF FLEXIBLE PAVEMENTS The availability of highly sophisticated computerized sohitions for multi- layered systems, coupled with recent advances in materials evaluation, has led to the development of several design methods that are based wholely or partly on theoretical analysis.*“" The more commonly used design methods are the Asphalt Institute method, the AmericanAssociation of State Highway and Trans- portation Officials (AASHTO) method, and the California method. These are presented in the following sections. Asphalt Institute Design Method In the Asphalt Institute design method, the pavement is represented as a multi- layered elastic system. The wheel load Wis assumed to be applied through the tire as a uniform vertical pressure po, which is then spread by the different compo- nents of tlie pavement structure and eventually applied on the subgrade as a much lower stress p,. This is shown in Figure 18.3. Experience, established theory, and test data are then used to evaluate two specific stress-strain conditions. The first, shown in Figure 18.3(b), is the general way in which the stress py is reduced to p, within the depth of the pavement structure and the second, shown in Figure 18.4, is the tensile and compressive stresses and strains imposed on the asphalt due to the deflection caused by the wheel load. Thickness design charts were -.developed, based on criteria for maximum tensile strains at the bottom of the asphalt layer and maximum vertical compressive strains at the top of the sub- grade layer. The procedure originally involved the use of a computer program, but the charts presented here can be used without computers. A detailed descrip- tion of the method is given in Thickness Design—Asphalt Pavements for High- ways and Streets.’ Design Procedure The principle adopted in the design procedure is to determine the minimum thickness of the asphalt layer that will adequately withstand the stresses that develop for the two strain criteria discussed earlier—that is, the vertical compres- sive strain at the surface of the subgrade and the horizontal tensile strain at the bottom of the asphalt layer. Design charts have been prepared for a range of traffic loads. This range is usually adequate for normal traffic volumes encoun- tered in practice. However, when this range is exceeded, the computer version should be used. The procedure consists of five main steps. 1. Select or determine input data. 2. Select surface and base materials. 3. Determine minimum thickness required for input data. : 4. Evaluate feasibility of staged construction and prepare stage construction plan, if necessary. 5. Carry out economic analyses of alternative designs and select the best design.DESIGN OF FLEXIBLE PAVEMENTS 795 Step 1: Design Inputs. The design inputs in this method are traffic charac- teristics, subgrade engineering properties, and subbase and base engineering properties. Traffic Characteristics. The traffic characteristics are determined in terms of the number of repetitions of an 18,000 Ib (80 kilonewtons (KN)) single-axle load applied to the pavement on two sets of dual tires, This is usually referred to as the equivalent single-axle load (ESAL). The dual tites are represented as two circular plates each 4.52 in. in diameter, spaced 13.57 in. apart. This representation corre- sponds to a contact pressure of 70 lb/in.?. The use of an 18,000 1b axle load is based on the results of experiments that have shown that the effect of any load on the performance of a pavement can be represented in terms of the number of single applications of an 18,000 Ib single axle. A series of equivalency factors used in this method for axle loads are given in Table 18.3. To determine the ESAL, the number of different types of vehicles such as cars, bises, single-unit trucks, and multiple-unit trucks expected to use the facility _ during its lifetime must be known. The distribution ‘of the different types of vehicles expected to use the proposed highway can be obtained from results of classification counts that are taken by state highway agencies at regular intervals. In cases where these data are not available, estimates can be made from Table 18.4, which gives representative values for the United States. When the axle load of each vehicle type is known, these can then be converted to equivalent 18,000 Ib loads using the equivalency factors given in Table 18.3. The equivalent 18,000 lb load can also be determined from the vehicle type, if the axle load is unknown, by using a truck factor for that vehicle type. The truck factor is defined as the number of 18,000 Ib sirigle-load applications caused by a single passage of a ~-vehicle, These have been determined for each class of vehicle from the expression truck factor = 2_(tumber of axles x load equivalency factor) number of vehicles Table 18.5 gives values of truck factors for different classes of vehicles; however, these factors are based on data collected prior to the deregulation of the trucking industry. Recent-analysis of limited data has shown that truck factors can be as high as 5.0, even though the maximum value given in-Table 18.5 is only 1.99. Therefore, itis advisable, when truck factors are to be used, to collect data on axle j loads for the different types of vehicles expected to use the proposed highway and to determine realistic values of truck factors from that data. The total ESAL applied on the highway during its design period can be determined only after the design period and traffic growth factors aré known. The design period is the number of years the pavement will effectively continue to carry the traffic load without requiring an overlay. Flexible highway pavements are usually designed for a 20-year period. Since traffic volume does not remain constant over the design period of the pavement, it is essential that the rate of growth be determined and applied when calculating the total ESAL. Annual growth rates can be obtained from regional planning agencies or from state796 ° DESIGN OF FLEXIBLE PAVEMENTS Tablé 18.3. Load Equivalency Factors Load Equivateney Load Equivalency Gross Axle Load oe Gross Axle Load ae Single Tandem Single | Tandem “a ® Axles Axles 2 ® Axles | Axles 445 | 1,000 | 0.00002 1825 | 41,000 | 2327 2.2 89 2.000 | 0.00018 170 | 42000 | 2564 251 1335 | 3,000 | o.oga7e 191.3 | ° 43000 -| 2022 275 178 4,000) 0.00209 195.7 | 44000 | 31.00 3.00 2225 | 5000 | 0.00500 200.0 | 45,000 | 34.00 327 26.7 600 | ‘o.01043 204.5 | 46000 | 37.24 3.55 316 | 7000 | o.0196 200.0 | 47000 | 40.78 | 3.85 35.6 8000 | 0.0343 2135 | 48000 | 44.50 an 400 9000 | ‘0.0502 2180 | 49000 | 4854 451 445 | 10,000 | 0.0877 | oooss | 2224 | 50,000 | s2ee 4.96 48.9 | 11,000 | 0.1311 | ootooa | 2268 | 51,000 523 saa | 12000 | ~-o11@9 corse | 231.3 | 52,000 563 57.8 | 13000 | 0.264 0.019 | 2357 | 53,000 6.04 623 | 14000 | » 0.360 0.0270 | 240.2 | 54,000 6.47 65.7 | 15000 | 07a 0.0360 | 244.6 | 55,000 6.93 71.2 | 16000 | 0.623 0.0472 | 249.0 | 56,000 7at 756 | 17000 | 0.796 0060s | 2535 | 57,000 798 00 | 18000 } 1.000 0.073 | 2580 | 58,000 845 ss | 19000 | 1.28 0.0971 | 2625 | 59,000 901 820 | 20000 | 1.61 0.1208 | 267.0 | 60,000 959 934 | 21,000 | 1.03 o1a8 271.3. | 61,000 10.20 978 | 22000 | 218 0.180 2758 | 62000 10.84 1023 | 23000 | 258 0217 280.2 | 63,000 1182 106.8- | 24,000 | 303 0.260 28.5 | 64,000 1222 11.2 | 25000 | 3.63 0.308 209.0 | 65,000 1296 1156 | 26,000 | 409 0.364 293.6 | 66,000 1373 1200 | 27,000 | 4.71 0.426 208.0 | 67,000 1454 1245 | 28000 | 5.29 0.495 3028 | 68,000 15.38 1200 | 29000 | 6.14 0572 307.0 | 69,000 16.26 133.5 | 30000 | 697 0.658 3116 | 70,000 17.19 1380 | 31,000 |} 7.68 0.753 3160 | 71,000 18.15 1423 | 32000 | 8.98 0.887 3200 | 72,000 19.16 1468 | 33,000 | 9.98 og7t 326.0 | 73,000 20.22 151.2 | 34000 | 1138 1.095 32,0 | 74,000 21.32 1657 | 35000 | 1260° | 123 333.8 | 75,000 22.47 160.0 | 36000 | 13.03 138 338.0 | 76,000 23.66 1645 | 37,000 | 1680 153 312.8 | 77,000 2ao1 162.0 | 38000 | 17.20 170 347.0 | 78,000 26.22 173.5 | 39,000 | 19.06 1.89 351.5 .| 79,000 22.58 1780 | 40,000 | 21.08 2.08 358.0 "| 80,000 2899 Note: KN converted to Ib are within 0.1 percent of Ib shown. Source: Reproduced from Thickness Design—Asphalt Pavements for Highways and Streets, ‘Manual Series No. 1, The Asphalt Institute, College Park, Md., September 1981DESIGN OF FLEXIBLE PAVEMENTS 797 — Table 18.4 Distribution of Trucks on Different Classes of U.S. Highways Percent Trucks Truck Class | interstate Rural | Other. Rurat Al Rural All Urban | All Systems Average | Range | Average | Range | Average | Range | Average | Range | Averege| Range Single-unit trucks 2axle, d-tire 39 |17-64| 58 |40-80) 47 j2366/ 61 |33-04| 49 | 26-67 axle, 6-tire to | 55) 11 [4-18] 10 | 416] 19 | 426] 11 | 6-20 Seaxle or more 2 fra} 4a fre] 2 [ra] 3 far] 3 fas Allsingle-units | 81 [30-71] 73 |so-08] so [36-77] 77 |55-04/ 63 |a6-e1 Multiple-unit trucks Seale a fenz | oor fers 1 | 3 1 fea} oa Jere 4-axle s |r] 3 fre} 4 | to] 4 fars} 4 | vt0 S-axle ormore’ | 43 |24-59/ 23 | e40| 36 |16-87| 18 | 5-37] 32 |15-56 All multiple-units] 49 |31-71] 27 |13-50] 41 |23-66] 23 | 644 37 | 20-67 All trucks 100 100 100 100 100 ‘Note: Compiled from data supplied by the Highway Statistics Division, Federal Highway Administration, ‘Including full-trailer combinations in some states. Souree: Reproduced from Thickness Design—Asphalt Pavements for Highways and Streets, Manual Series No. 1, The Asphalt Institute, College Park, Md,, September 1981. * highway departments. These are usually based on traffic volume counts over several years. It is also advisable to determine annual growth rates separately for trucks and passenger vehicles since these may be significantly different in some cases. The overall growth rate in the United States is between 3 percent and 5 percent per year, although growth rates of up to 10 percent per year.have been suggested for some interstate highways. Table 18.6 shows growth factors for different growth rates and design periods, which can be used to determine the total ESAL over the design period. The portion of the total ESAL acting on the design lane is used in the determination of pavement thickness. Either lane of a 2-lane highway can be considered as the design lane, whereas for multilane highways, the outside lane is considered. The identification of the design lane is important because in some cases more trucks will travel in one direction than in the other or trucks may travel heavily loaded in one direction and empty in the other direction. Thus, it is necessary to determine the relevant proportion of trucks oni the design lane.798 DESIGN OF FLEXIBLE PAVENIENTS Table 18.5 Distribution of Truck Factors on Different Classes of U.S. Highways Truck Factors Rural Systems Urban Systeme A Systems T Interstate Rural Other Rural ‘All Rural Al Urban ‘Average | Range | Average | Range Range | Averoge ‘Average | Range Single-unit trucks Zarle Ate | 0.02 001-006 | 002 |0.01-009] a03" |0.02-0.08 | 0.03" | 001-005 | 002 | 001-007 Pante,tire | 0.19 J0.13-030] 021 |or4038]020 fo-oa1}o2s |ose-042| 021 |015-0:32 sonigormore | 086 |o0e-155] 073 Jastrsr|osr foz1s0|103 |oszrs0| 072 |aze150 Asing-wnis | 007 [002-016] aor [oo2ns7] oor {o00-018]002 | o0¢021 | o07 |ao2037 a _ ast foaonee | 047 Jozoe|oa [oss-aso loa forsio2| oe |ose-a76 dante 0.62 | 0.40-1.07 | 0.83 | 0.44-1.55 | 0.70 0.37-1.34 | 0.89 0.60-1.64 | 0.73 043-132" ‘S-axle or more” 0.94 Jaert.16 0.98 | 0.58-1.70 | 0.95 0.58-1.64 | 1.02 0.69-1,69 | 0.95 | 0.63-1.53 All muttipleunits | 0.93 |0.67-198} 0.97 |0.67-1.80}09% |as6-1.43 {1.00 [0.721.588 | 0.96 | 0.71-1.30 All trucks 049 }0:38-077| 031 Jo200s2}042 |o20-067}030 | 015-059] 0.40 | 027-069 ‘Note: Compiled from data supplied by the Highway Stati Administration. "Including full-trailer combinations in some states. ‘**For values to be used when the number of heavy trucks is low, see original source. Source: Reproduced from Thickness Design—Asphalt Pavements for Highways and Streets, Manual Series No. 1, The Asphalt Institute, College Park, Md., September 1981 Division, Federal Highway When data are not available to make this determination, percentages given in Table 18.7 can be used. The procedure for determining thé design ESAL is demonstrated in Examples 18-1 and 18-2, a Example 18-1 Computing Accumulated Equivalent Single-Axle Load for a : Proposed 8-Lane Highway Using Load Equivalency Factors An 8-lane divided highway is to be constructed on a new alignment. Traffic volume forecasts indicate that the average annual daily traffic (AADT) in both directions during the first year of operation will be 12,000, with the following vehicle mix and axle loads. . Passenger cars (1000 Ib/axle) = 50 percent 2-axle single-unit trucks (5000 Ib/axle) = 33 percent 3-axle single-unit trucks (7000 Ib/axle) = 17 percentDESIGN OF FLEXIBLE PAVEMENTS 799 — Table 18.6 Growth Factors _ 3! i ‘Anoual Growth Rate, Percent) Period, ; No : yeast) | gi, | 2 4 5 6 7 8 10 1 10 | 10 | 10 | 1 | 1 | so | 10 2 20 | 202 | 204 | 205 | 200 | 207 | 208 3 30 | 306 | ai | ais | 313 | a2 | 325 4 40} a2 | 425 | ast | aay | ae | ar 5 80 | s20 | saz | sss | see | 7s | 587 6 so | 631 | 66s | 60 | os] 715 | 724 7 70 | 743 | 790°} a1 | aso | aes | ase 8 go | ase | 21 | oss | 900 | 1025 | 1064 9 90 | 975 | 1088 | 1103 | 1140 | 1198 | 1240 10 yoo | 1095 | 1201 | 1288 | sate | i282 | 1449 n 110 | 1217 | 1349 | rear | 97 | 1578 | 1065 2 120 | 1941 | 1503 | 1502 | 1687 | 1709 | 1098 8 130 | ta6e | 1663 | 1771 | 1988 | 2014 | 21.80 14 140 | 1597 | 1223 | 1916 | 2101 | 2255 | 2421 15 150 | 17.29 | 2002 | 2188 | 2za2e | 2519 | 2715 16 160 | 1864 | 2162. | 2366 | 2567 | 2709 | 2032 7 120 | 2001 | 2370 | 2584 | 2021 | soe | 2375 18 sao | 2a1|. 2565 |. 2813 | sao | a400 | 3745 19 190 | ze4 | 267 | 3058 | 3376 | 2738 2 zo | 2430 | 2078 | 3308.| 3670 | 21.00 25 260 | 3203, | 4165 | 4773°| S96 | 0325 20 soo | 4057 | 608 | coca | 7906 | e446 18449 35 | 350 | 4a90 | 73.65 | 9032 | unnaa | 13024 zn02 Note: Factor = [(1 +1)"— I]/r, where r = and is not zero. If annual growth is zero, growth factor = design period. Source: Reproduced from Thickness Design—Asphalt Pavements for Highways and Streets, Manual Series No. 1, The Asphalt Institute, College Park, Md., September 1981. The vehicle mix is expected to remain the same throughout the design life of the pavement. If the expected annual traffic growth rate is 4 percent for all vehicles, determine the design ESAL, given a design period of 20 years. ~ A general equation for the accumulated ESAL for cach category of axle load is obtained as ESAL, = fy x Gy x AADT, x 365 N, x Fei where ESAL, = equivalent accumulated 18,000 Ib (80 KN) single-axle load for the “axle category i Su = design lane factor G;, = growth factor for a given growth rate j and design period ¢ AADT, = first year annual average daily traffic for axle category N, = number of axles on each vehicle in category i Fe, = load equivalency factor for axle category i