A Critical Review of Pavement Design Methods Based on a Climate Approach

Abstract

The design of flexible road pavements is a complex process as a result of the multiple variables that influence and interact in the models that allow the design of each layer. In recent years, a particular interest has been raised to ensure that climate is considered in pavement design due to temperature and precipitation that influence the deterioration of pavements, impacting their service life. This paper presents a critical review of flexible pavement design methods, from the first ones based on experience, such as empirical methods, to the most recent ones on mechanical–empirical methodologies, where, based on different principles, they determine the thicknesses of the layers that integrate the structure of a pavement to identify how these methods have included climate variables within their methodology. Through this review, it was identified that temperature is incorporated in the dynamic modulus of the asphalt mix, and precipitation and moisture are incorporated through the resilient modulus in the granular layers (base, subbase, and foundation soil or subgrade courses). As a result, it was identified that the most holistic way of integrating climate is through the Enhanced Integrated Climatic Model (EICM) from the Mechanistic–Empirical Pavement Design Guide (MEPDG). In many cases, climate is incorporated through parameters whose behavior is associated with temperature and precipitation but does not use the data of these climate variables directly from the project site. The practical incorporation of climate into design methods allows an increase in the certainty of results, ensuring additional climate-resilient pavement structures and increasing their durability and sustainability during their service life.

1. Introduction

The importance of pavements, beyond the functional and practical benefits for road users, is that they are a fundamental asset of transportation systems, but they are also a basic component of the functioning of the societal system [1]. Thus, pavements are considered the most important asset of road infrastructure in transportation systems.

Pavement design might be defined as the determination of the thickness of pavement, which has to be placed over a soil formation in a particular environment (the natural conditions) to provide a satisfactory riding surface for a given set of external conditions [2].

To ensure pavement durability, various institutions have developed several methods with empirical and mechanistic–empirical approaches to model pavement performance and deterioration as a function of different variables [3]. For flexible pavements, methods have been based on soil behavior, the use of theories, statistical analysis of field tests, etc., which have made it possible to classify the methods and study them [4,5,6].

The main variables that influence pavement design and performance are shown in Figure 1. It details how these variables interact with each other, which makes it a complex problem [7]. The variables considered are pavement structure design, traffic design, pavement construction, pavement maintenance, and the environment. These factors, all together, determine the service life of pavement.

Corro & Prado [8] expose that the main variables involved in the design of a flexible pavement can be classified into three categories:

(a)

Structural. These include characteristics related to each of the layers that compose the road pavement, such as thickness, strength, and deformability in the expected service conditions. This relates to the materials that conform to the structure and the stresses required for the design.

(b)

Traffic load. This refers to the effects produced by mixed traffic as it travels along the road. In this case, data related to Annual Average Daily Traffic (AADT), annual growth rate, single or tandem axle loads, histogram of traffic distribution in the road section, and pavement design life before the road requires reconstruction in years, in which case the pavement failure criterion must be defined in advance. The mixed traffic should be transformed into equivalent traffic on single axles using theoretical-empirical factors. New approaches use load spectra to characterize traffic more accurately.

(c)

Climate and regional conditions. The rheological characteristics of the materials that constitute the road depend on the temperature, precipitation regime, average annual precipitation, water table, geology, and topography of the region.

Under the empirical methodology, the pavement design seeks to determine the thickness of the layers according to the material that will be used to support the loads of vehicles traveling based on previous experiences and observations of their behavior in field studies [9]. Empirical methods describe variables in climate where the design procedure does not always clearly define how they should be considered, which are currently of special interest since climate factors influence the physical, mechanical, and chemical characteristics of materials. For example, temperature (influences on solid state, thermoelastic, heat conservation and transfer, freezing, and thawing) [10,11,12,13], precipitation regime (mean and maximum annual precipitation, influences on relative moisture) [14,15], water table (influences in capillarity in the sublayers of the pavement) [16], solar radiation (impacts the destruction of asphalt bonds by UV light) [17,18], among others.

Environmental conditions have a significant effect on pavement performance [19]. Climate influences the rate of pavement deterioration, and, therefore, pavement maintenance and lifecycle costs [20]. Therefore, environmental factors and their variations are fundamental elements to be considered in pavement design and maintenance [21].

The environmental factors affecting pavements can be classified into two categories: external and internal. External factors that have a particular influence on the performance and durability of a pavement are temperature and precipitation, water table, and freeze-thaw cycles. The internal factors that impact pavement performance are moisture, drainage in the layers, and infiltration [19]. There are other external factors associated with the road design that influence the durability of the pavement, such as the road width, since a larger surface area requires better surface drainage, a cross slope to remove excess water rapidly from the road surface, the type of pavement and its condition, and ditch bottom width, but these elements are not well considered in pavement design.

To evaluate how climate has been considered, a review of historical design methods and new design approaches was conducted and then analyzed in detail to further identify how some of them have considered and incorporated climate into the structural design of a flexible road pavement. This review focused on methods used in North America (including Mexico) and those where an English version of the guide was available. Some pavement design manuals were identified, which did not include the methodology, so these are not part of this paper. Through this review, findings were identified to propose a practical way to incorporate climate into pavement design.

2. Review of Flexible Pavement Design Methods

2.1. Early Pavement Design Methods

The California Highways Department developed a method for determining the thickness of a pavement based on the California Bearing Ratio (CBR) test [6,22]. The method establishes that the penetration resistance of the base material can be taken as a reference value for pavement design and could support axle loads in an economical manner, which are classified into three groups according to traffic conditions: light, medium, and high. Using the CBR of the base and the wheel load, the thickness of the layers is obtained through charts.

The Missouri State Highway Department used an empirical method for the design of flexible pavements, considering two factors: the “Group Index” and “Heavy Vehicles” [23]. The group index is not used to place the soil in a particular group but rather as a means of rating the value of the soil as a subgrade material within its own group. The group index of the soil and the daily traffic allow for obtaining the thickness of the layers through charts.

The limiting shear failure method is used to determine the thickness of pavements so that shear failure does not occur. A first approach was made by Barber [24], in which Terzaghi’s theory was applied [25] to determine pavement thicknesses. To introduce this theory, Barber applied the results of the triaxial compression test to unaltered soil samples. For his part, McDowell [26] constructed a diagram based on the triaxial test procedure developed at the Texas Highway Department, which consists of Mohr envelopes that have been correlated with road service. After classifying the soil according to its stresses (six different classes) and the wheel load, the thickness of the pavement base can be determined.

These early methods considered soil properties as a first effort to characterize the environmental behavior of the site where pavement is to be designed and constructed.

A limit deflection method is used to determine the thickness of a pavement while considering that the allowable limit of vertical deflection should not be exceeded. Palmer & Barber [27] developed theoretical expressions based on the Boussinesq displacement for an elastic and homogeneous layer, which made it possible to determine the thickness of a pavement. The State Highway Commission of Kansas modified the Boussinesq equation and limited the value of the deflection [28]. The reference equation included a saturation coefficient as a function of average annual precipitation, which influences the determination of the required pavement thickness (see Section 3).

The French catalog of standard pavement structures used two criteria for designing these structures: a limit for pressure on the subgrade, inferred from existing CBR charts, and a limit for strain in the bound courses, estimated from deflection measurements at the pavement surface [29].

An elasticity approach is the Shell method to determine the pavement thickness through charts, where the pavement structure is represented as a linear elastic multilayer system, in which the material is characterized by Young’s modulus of elasticity and Poisson’s radius. Traffic is represented as a load acting vertically and horizontally on the surface [30,31]. The method incorporates the elastic modulus of the subgrade as one of the main design parameters determined by representative vehicle loads and equilibrium moisture content. Based on these modules, the thickness of the binder course can be determined according to the type of asphalt mix and the temperature of the site.

The methods that are based on statistical analyses, which allowed the development of regression equations and charts for pavement design, were supported by the results obtained in multiple experimental tests in sections in the USA, developed by the American Association of State Highway Officials (AASHO) and American Association of State Highway and Transportation Officials (AASHTO) [32,33,34] and the National Autonomous University of Mexico (UNAM) in Mexico [8], as well as a modification carried out by the Asphalt Institute [35]. The main feature of these US methods is the Present Serviceability Index (PSI) as a measure to evaluate the surface condition of the pavement derived from the traffic of loaded vehicles, reflected in the longitudinal and transverse roughness, the introduction of the effective resilient modulus of the foundation soil or subgrade, and the modulus of elasticity as a fundamental property of any material that conforms the pavement layers.

The main disadvantage of empirical methods is that they can only represent a limited set of material conditions, loading, and even site environmental considerations, which vary seasonally and affect the reliability of the results.

2.2. New Approaches for Pavement Design

Advances in technology have prompted the development of new approaches or the adaptation of existing methods for the design of flexible pavements. These include two main concepts: design based on a mechanical analysis, where alternatives are evaluated through the mechanical properties of the pavement structure, such as stresses, deflections, etc.; and performance model-based design with a mechanistic–empirical approach, which is developed using mechanistic models that combine field studies [36].

The AUTh methodology for flexible pavement design is a semi-analytical procedure originally developed by a group of scientists [37]. The AUTh pavement design methodology includes charts and tables that allow determining the thickness of all flexible pavement layers consisting of dense asphalt concrete with asphalt of penetration grade (40/50 or 60/70) and unbound materials for base/subbase layers [38]. The base and subbase are treated as a unified layer. The thickness of this layer can be 200, 300, or 400 mm, which simplifies the construction work (the base/subbase is constructed in 100 mm thick layers). A similar approach uses the standard to design new pavement construction for the United Kingdom using four different types of foundations, and then, with the number of millions of axles, the thickness of the pavement is determined. The thickness of the binder course is determined according to the type of cementitious material, either AC 40/60 or EME2, using charts [39].

A method for the rational design of pavement structures was developed in France that does not calculate the thicknesses of a pavement based on the projected service life but rather estimates the cumulative damage over time of a proposed pavement structure. The design catalog provides thicknesses of each layer with respect to subgrade moduli of 20, 50, 120, and 200 MPa and standard cumulative axle categories [40].

In 2008, AASHTO introduced a new approach to pavement design called the “Mechanistic–Empirical Pavement Design Guide (MEPDG)” [41]. Two- and three-layer pavements based on mechanistic methods focus on measurable structural responses calibrated to the in-service behavior of asphalt pavements, such as superficial deflection or vertical stress. Mechanistic structural design is based on two main deterioration models: permanent deformation and fatigue.

Due to the amount of data, mechanistic methods use different software. For example, Austroads calculates load response using linear theory and, specifically, the CIRCLY program [42], the French rational method is based on the ALIZE program (version 1.5) [43], MEPDG calculates critical pavement responses using the elastic layer theory program identified as JULEA integrated in AASHTOWare [41], and stresses and strains in pavement layers in India are analyzed using the IITPAVE software used in the Indian Road Congress method [44]. The California Department of Transportation (CalTrans) uses CalME, which is a computer program that uses an “Incremental Recursive” (I-R) approach that models the entire damage process [45]. The Mexican Institute of Transportation developed a design method, the IMT-Pave program, with a mechanistic approach [46], considering the pavement as a multilayer structure where the behavior of the materials is based on the Theory of Elasticity, and the National Laboratory of Materials and Structural Models (LANAMME) in Costa Rica developed a mechanistic model for pavement design (CR-ME) [47], based on incremental damage and applying Miner’s Law concept.

Mechanistic–empirical methods are based on the stratified elastic theory, which models the structure as a semi-infinite continuum, divided into layers of finite thickness over an elastic half-space [48]. In addition, other alternatives to the elastic theory have been developed, such as the use of Odemark’s equivalent layer thickness [49], which allows for quick calculation of the responses by transforming a multilayer medium into a single layer with a transformed thickness [50]. Furthermore, the probabilistic stress distribution allows the prediction of stresses in flexible pavements with the central probability limit theorem and a lateral stress coefficient for each material [51]. Additionally, artificial neural networks have been used to analyze the structure of the flexible pavement and determine its critical responses under the influence of a standard axle load [52]. Finite element models have been widely applied to the design and analysis of pavement structures. Three types of models have been used to study multilayer pavement structures: plane strain, axisymmetric, and three-dimensional (3D) [53].

Most mechanistic or mechanistic–empirical methods have the possibility of considering climate as part of the design variables, through variables such as precipitation and temperature, or implicitly within the material properties.

3. Climate Considerations in Pavement Design Methods

Climate serves as an essential input in pavement design, and depending on its variability, it can have a significant impact on pavement performance [53]. However, climate is considered in several cases implicitly within the material properties, and only in the most recent methods explicitly.

3.1. First Approaches That Included or Considered Climate in Pavement Design

The State Highway Commission of Kansas (1947) proposed the use of the following formula to determine the required pavement thickness, considering the soil saturation coefficient as a function of average annual rainfall [28]. Equation (1) for determining the required pavement thickness is as follows:

$$T=\left[\sqrt{{\left[{\displaystyle \frac{3Pmn}{2\pi CS}}\right]}^2-a^2}\right]\left[\sqrt[3]{{\displaystyle \frac{C}{Cp}}}\right]$$

(1)

where T is the thickness required (in), Cp is the modulus of the pavement or surface course (psi), C is the modulus of deformation of subgrade or subbase (psi), p is the base wheel load (lbs), $a$ is the radius of area of tire contact corresponding (in), S is the permitted deflection of surface (in), m is the traffic coefficient based on volume of traffic (see

Table 1. Traffic coefficient [28].

Traffic Coefficient (m)	Wheel Load (lb)	Total Traffic (Veh Per Day)
1	9000	>1500
5/6	7500	900–1500
2/3	6000	300–900
1/2	4500	50–300

), and n is the saturation coefficient based on rainfall (see

Table 2. Saturation coefficient [28].

Saturation Coefficient (n)	Average Annual Rainfall (in)
1.0	35.0–45.0
0.9	30.0–34.9
0.8	25.0–29.9
0.7	20.0–24.9
0.6	15.0–19.9

).

The saturation level was obtained from different on-road records to determine the values of the saturation coefficient to be used in the method. Table 1 shows the values for the saturation coefficient.

The AASHO method (1961) considered a regional adjustment factor due to climate since it had been detected that the climate and foundation soil conditions of the tests performed corresponded to a particular site, and, therefore, the results were only valid for that site [32]. The regional factor varied from 0.5 to 5, as shown in Figure 2, depending on whether the foundation soil is frozen, dry, or saturated.

The approach to considering the climate variations in the foundation soil was through the soil-bearing capacity and the regional factor. However, it considered a single value for the determination of the thicknesses and, therefore, failed to consider the seasonal variability.

The updated AASHTO method (1986) considered that the environment can affect pavement performance in different ways [33]. The guide introduced the “Effective Resilient Modulus” of the roadbed soil (foundation or subgrade soil). The purpose of identifying the seasonal modulus is to quantify the relative damage to which a pavement is subjected during each season of the year, as shown in Figure 3.

The AASHTO guide also includes a “Drainage Coefficient (m)” since water drainage in pavements is an important consideration in roadway design. The excess water in the pavement layers, particularly those composed of granular materials, combined with the stresses due to traffic can cause premature damage to the pavement structure [33,34]. The drainage coefficients are values less than, equal to, or greater than one, depending on the percolating quality of the layers, which are illustrated in

Table 3. Recommended Drainage Coefficient “m” values for modifying structural layer coefficients of untreated base and subbase materials in flexible pavements [33].

Quality of Drainage	Percent of Time Pavement Structure is Exposed to Moisture Levels Approaching Saturation
	Less than 1%	1–5%	5–25%	Greater than 25%
Excellent	1.40–1.35	1.35–1.30	1.30–1.20	1.20
Good	1.35–1.25	1.25–1.15	1.15–1.00	1.00
Fair	1.25–1.15	1.15–1.05	1.00–0.80	0.80
Poor	1.15–1.05	1.05–0.80	0.80–0.60	0.60
Very Poor	1.05–0.95	0.95–0.75	0.75–0.40	0.40

, according to the AASHTO proposal [33].

The Asphalt Institute’s method “MS-01” (1970) considered that environmental conditions adversely affect the bearing properties of subgrade materials. The three main critical parameters that affect bearing properties are moisture, soil expansion, and freezing. These can be represented by the bearing capacity ratio (CBR) and the resistance value (R-Value) [35]. The design value of the subgrade is 90% of all values obtained from the CBR test of the subgrade soil at the design site following the MS-01 manual. The result is obtained according to

Table 4. Percent of CBR values equal to or greater for each different value (based on the example of the Asphalt Institute) [35].

CBR	Number Equal to or Greater Than	Percent Equal to or Greater Than
6	11	(11/11) 100 = 100
7
7	10	(10/11) 100 = 90.9
8	8	(8/11) 100 = 72.7
9
9	7	(7/11) 100 = 63.6
10
10	5	(5/11) 100 = 45.4
11
11	3	(3/11) 100 = 27.3
12	1	(1/11) = 9.1

, whose calculated percentages are plotted as shown in Figure 4.

The UNAM method used in Mexico considers that climate conditions influence the bearing capacity of the soil (VRS), so the relative value of critical support for design considers a coefficient of variation that includes the uncertainty due to climate. To determine this value, it is recommended that the road be mapped to consider the climatological and geotechnical conditions to adequately regionalize the climate zones of the country and, thus, estimate the critical value of the VRS [55].

3.2. Incorporating Climate Factors in Pavement Design

The Asphalt Institute’s method updates have considered the resilient modulus to replace the CBR and the incorporation of temperature for the determination of asphalt concrete layer thickness. The method for simulating the effects of temperature over time (seasonal variation) was based on a study relating modulus–temperature and asphalt properties. The charts for determining the thicknesses consider three types of climate and temperature ranges. Charts from the Asphalt Institute were adopted by the Central American Economic Integration Secretariat [56]. The 1999 version of the manual considers a broader spectrum of temperatures [57].

The Shell method considers temperature variation, which does not have a significant effect on the resilient modulus of granular materials but has a strong influence on the properties of asphalt [31]. To incorporate the temperature, a parameter is introduced, “Weighted Mean Annual Air Temperature” (w-MAAT), which is obtained from the monthly average air temperature for a specific site. Average temperature information is generally obtained from meteorological stations. The w-MAAT parameter is obtained from the chart in Figure 5.

For each monthly average air temperature, a weighting factor is obtained, which is arithmetically averaged to later obtain the w-MAAT. This calculation is supported by the worksheet shown in

Table 5. Determination of weighted air temperature (worksheet based on the Shell method) [31].

Month	MMAT, °C	Chart W: Weighting Factor
January	8	0.21
February	8	0.21
March	12	0.36
April	16	0.62
May	19	0.93
June	22	1.40
July	26	2.35
August	28	3.00
September	22	1.40
October	19	0.93
November	12	0.36
December	6	0.16
Total of weighting factors		11.93
Average of weighting factor		say 1.0
Chart W: w-MAAT, °C:		19.5 say 20

.

The value of w-MAAT is used in charts to determine the asphalt layer thicknesses, depending on the thickness of the base and the modulus of the subgrade, the type of asphalt mix (eight types of mixes are considered), the number of axles accumulated during the service life (80 kN axles), and temperatures (4, 12, 20, and 28 °C). Figure 6 shows an example of charts for determining the thickness of the asphalt layer.

The French method considers in its design the following environmental information as relevant: the hydric condition of the support soil, seasonal temperature cycles, and the intensity of frost periods [40]. The hydric condition of the support soil establishes the use of the most unfavorable situation at the project site. The French design method recommends the use of Miner’s rule to calculate the equivalent temperature to consider the seasonal cycles of temperature and its sensitivity to the deformability and strength properties of asphalt materials. In relation to frost, it uses an atmospheric frost index chosen as a reference (IR) and an allowed frost index (IA).

AASHTO’s Mechanistic–Empirical Pavement Design Guide (MEPDG) establishes a climate/environmental analysis, including temperature and precipitation as environmental variables [58]. The approach to incorporating climate information into the MEPDG design is through the Integrated Climate Model. This program was recognized as the most comprehensive model that addresses the effects of climate on pavements [59]. The original version was analyzed and adapted [60,61] and called “Enhanced Integrated Climatic Model Version” (EICM), which has since been updated, calibrated, and validated in different research studies [62,63].

The EICM allows for predicting or simulating behavioral patterns of pavement and material characteristics, along with environmental conditions of various climate variables that are used in the MEPDG design method. To estimate the resilient modulus at equilibrium ($M_{req}$) of the foundation soil or subgrade, an environmental factor ($F_{env}$) is obtained from the EICM, as shown in Figure 7, and multiplied by the optimum resilient modulus ($M_{ropt}$).

The MEPDG climate module calculates temperature and moisture conditions throughout the pavement structure on an hourly basis, i.e., the temperatures of each HMA (Hot Mix Asphalt) layer are combined into five quintiles (five successive groups of 20% each of the calculated values) for each month of the period of analysis of load-related deterioration. The frequency distribution of the HMA temperatures used is assumed to have a normal distribution. The average temperature within each quintile of a layer for each month is used to determine the dynamic modulus. Dynamic modulus is a measure of the stiffness of asphalt mixtures under dynamic (or cyclic) loading conditions. The dynamic modulus decreases with increasing temperature; thus, in hot climates, lower dynamic modulus values can lead to increased rutting and deformation under heavy traffic loads, and in cold climates, higher modulus values can increase the risk of thermal cracking.

The CalME (Caltrans Mechanistic–Empirical) method uses the EICM as the AASHTO MEPDG, using a previously calculated external database, which makes the design process faster. CalME calculates sub-surface temperatures using the one-dimensional finite element method [65].

4. Analysis of Pavement Design Methods

4.1. Climate-Related Factors Used in Pavement Design Methods

Climate-related factors, especially temperature and precipitation, significantly affect the quality and lifespan of roads [66].

As a result of the analysis of the design methods and, specifically, how they consider the climate of the site to evaluate its effects on the different pavement layers, the mechanical property by which climate and its variability can be considered in the current design methods, as well as future adaptations of the mechanical–empirical method for pavements, were identified for each of the layers.

For the asphalt layer, it is important to consider the stiffness behavior of the asphalt. This property will have different proportions, which depend on the loading duration and the temperature at which the load is applied [67]. The dynamic modulus is used to calculate the horizontal and vertical deformations of an asphalt mixture and allows for determining the maximum permanent deformation within each layer and the location of the maximum fatigue damage, where there is a strong relationship between modulus and temperature.

The base and subbase materials are used as a drainage layer and to control the capillary rise of water, which protects the pavement structure, which, therefore, generally uses granular materials (unbound soils). The monthly average moisture content relative to the optimum moisture content should be used to adjust the resilient modulus of each unbound layer to consider variations in the modulus as a function of water content.

AASHTO [33] introduced the effective resilient modulus of the foundation soil or subgrade, which variable allows estimating a single weighted value of the seasonal modulus since weather conditions influence the bearing capacity of the soil.

The moisture content of soil has a strong impact on the resilient modulus value, as it decreases with increasing moisture content.

4.2. Key Findings from the Analysis of Pavement Design Methods

A critical review of pavement design methods shows that they do not adequately incorporate climate (precipitation and temperature), except for MEPDG, which does it through EICM.

For example, to incorporate the effect of precipitation, it is performed implicitly through the resilient modulus for granular soils but does not consider its seasonal variation. Only AASHTO (86 and 93 versions) allows for obtaining an effective resilient modulus for the foundation soil or subgrade, but it requires multiple field measurements to be determined, which makes the projects more expensive, and, therefore, the necessary number of tests are not performed. In some methods, they suggest measuring in the most unfavorable conditions of the year.

Regarding temperature, which was considered in both the Asphalt Institute method and the Shell method, it was determined that the temperature is considered directly or weighted to determine the thickness of the asphalt layer. The weighted temperature allows for considering the seasonal variation in the temperature, while the MEPDG transforms the temperatures into quintiles to determine the properties of the asphalt mixes.

Climate has been incorporated into the AASHTO MEPDG, including the estimated temperatures and precipitation for the site according to historical weather station data and with support from AASHTOWare.

As a result of the analysis, the EICM implementation of the MEPDG can continue to be an excellent input tool for incorporating climate into design methods; however, the expected future climate variation under some selected climate scenario for the project region must be considered. The historical climate information can be incorporated in the EICM through the Thornthwaite Moisture Index (TMI), which allows calculating the suction matrix and the degree of saturation obtained from the soil characterization of the project site but is practical only for granular soils (unbound materials) to be used as foundation soil and subgrade, but it can also be used for the granular materials but used in base and subbase layers.

The EICM cannot be easily applied in any region because it requires that a database with weather information be developed, calibrated, and validated so that the EICM models can be used to feed the MEPDG deterioration models systematically. Validation of the EICM model is critical for pavement design and will either lead to over-design, resulting in high construction costs, or under-design, resulting in premature pavement failure [68].

Table 6. Climate key findings identified from the critical review of pavement design methods.

Method	Design Principle	How Is Climate Included?	Key Factor
State Highway Commission of Kansas (1947) [28]	Limit deflection method	Consider soil saturation coefficient as a function of average annual rainfall.	Modulus of deformation of subgrade or subbase
AASHO method (1961) [32]	Regression methods based on road test	The structural number is adjusted for a regional factor due to climate.	Weighted Structural Number
AASHTO method (1986, 1993) [33,34]	Regression methods based on road test	The guide introduced the “Effective Resilient Modulus” of the roadbed soil, considering seasonal modulus to quantify the relative damage subjected to moisture changes.	Subgrade Resilient Modulus
AASHTO method (1986, 1993) [33,34]	Regression methods based on road test	Also includes a “Drainage Coefficient” for modifying structural layer coefficients of untreated base and subbase materials in pavements.	Structural layer coefficients
Asphalt Institute’s method (1970) [35]	Limit elastic strains	Consider environmental conditions that adversely affect the bearing properties of subgrade materials.	Subgrade bearing capacity ratio (CBR)
UNAM method (1974) [8,55]	Regression methods based on road test	Considers that climate conditions influence the bearing capacity of the soil (VRS).	Subgrade bearing capacity ratio (CBR)
Asphalt Institute’s method (1999) [57]	Limit elastic strains	Simulate the effects of temperature over time (seasonal variation) was based on a study relating modulus-temperature and asphalt properties. The temperature is used in charts to determine the asphalt layer thicknesses	Dynamic modulus of HMA
Shell method (1978) [31]	Limit elastic strains	Considers temperature variation in the properties of asphalt. Introduced “Weighted Mean Annual Air Temperature” (w-MAAT). The value of w-MAAT is used in charts to determine the asphalt layer thicknesses.	Dynamic modulus of HMA
LCPC & SETRA (1994) [40]	Rational design	Considers in its design environmental information as relevant: the hydric condition of the support soil, seasonal temperature cycles, and the intensity of frost periods.	Soil elastic modulus
AASHTO (2008) [41]	Mechanistic–Empirical	Include temperature and precipitation as environmental variables through the EICM, using the TMI.	Subgrade Resilient Modulus
AASHTO (2008) [41]	Mechanistic–Empirical	Temperatures of each HMA layer are combined into five quintiles. The average temperature within each quintile of a layer for each month is used to determine the dynamic modulus	Dynamic modulus of HMA

shows a summary of the key findings identified from the critical review of pavement design methods to identify how these have incorporated climate and which properties are correlated with these considerations.

The methods in Table 6 have their origins in the USA, except for UNAM (Mexico), Shell (UK), and LCPC & SETRA (France). The use of these methods, in general, has had a worldwide impact, particularly the AASHTO methods. The Shell method and the Asphalt Institute method have been widely promoted by asphalt companies worldwide. The UNAM and LCPC&SETRA methods are widely used in the country of origin.

The factors associated with road design (width, cross slope, etc.) are not considered in pavement design methods, even though these have some influence on the durability of the pavement.

In all cases, the pavement design methods use typical historical weather patterns that reflect the local climate and, therefore, do not consider climate change, which has significant effects on the performance of a pavement.

If climate change is not considered in pavement design, it can accelerate the degradation of the structure, the materials that constitute the layers, and the foundation soil [69].

4.3. Climate Change in Pavement Design Methods

The current pavement design methods are based on the latest or historical climate data. Therefore, it is insufficient to consider only the information on climate data that is used in the deterioration models that use mechanistic–empirical methods or material properties used in empirical or semiempirical methods. This standard approach to using climate information fails because it does not consider climate change in pavement performance models in design methods.

In relation to the importance of considering climate change, this has been expressed in some studies where they have addressed in different ways the estimated changes in climate or predicted changes based on different climate scenarios associated with the radiative forcing from greenhouse gases in the atmosphere on the performance of pavements [70,71,72,73] and how to incorporate them into design methods.

According to the investigations, changes in temperature and precipitation have been incorporated into the AASHTO MEPDG by first including the estimated temperatures and precipitation for the site according to historical weather station data and with support from AASHTOWare and then adding the projected temperatures and precipitation according to the climate scenarios or changes expected to understand how climate change impacts pavement performance [74,75,76,77].

The potential risk of climate change raises interest in how it may affect the deterioration rates of flexible pavements and how pavement life service would be reduced as a consequence [78].

Incorporating climate change is difficult as temperature and precipitation projections are different for each region and require climate projections based on climate scenarios obtained from general circulation models based on the different expected future socio-economic pathways.

4.4. Relevance of This Review

The critical review of the design methods is to highlight the importance of considering climate as a fundamental factor. An appropriate design ensures that the pavement has a long service life and requires less maintenance, which reduces costs in the long term and optimizes the use of materials and resources, which can contribute to sustainability.

Pavement designers will be able to identify uncertainty in the determination of layer thicknesses by not correctly considering the climate of the project site in empirical methods. Therefore, they should explore alternatives that allow for characterizing the properties of the site materials through their modulus or migrate toward mechanistic methods that offer a better alternative to considering the climate of the project site.

It is important that, as part of the pavement design, the road design be considered, since elements such as width, cross slope, and ditch bottom, which are part of the road drainage system, are influenced by the climate and have an impact on the durability and performance of the pavement.

During pavement design, ensure that the climate conditions of the project site are considered as part of the design process, which will significantly increase the performance and durability of the pavement.

5. Conclusions

As a result of the analysis, the design methods for flexible pavement show that the main concern was to model and represent the impact of the load of the vehicles and its effect on the pavement to ensure its durability, leaving behind the influence of other external factors, such as climate.

The critical review of the design methods showed the level of consideration that climate variables have in the design of flexible pavements, which in many cases is incorporated through parameters whose behavior is associated with temperature and precipitation but do not use the data of these climate variables directly from the project site. The exception is the MEPDG, which, through the EICM, estimates the TMI at each site, which is a function of precipitation and temperature.

The mechanical properties most commonly used in pavement design and which are influenced by climate are the dynamic modulus of the asphalt layer, the resilient modulus of the granular layers (base and subbase), and the foundation soil (subgrade). Therefore, climate should be incorporated into pavement design methods and the mechanical properties of materials, and each region/country should consider its own methodology for addressing local climate information, which is different for each project site.

Methods that currently consider climate in pavement design use historical data, but because pavements are designed for lasting periods beyond 20 years, it is necessary to consider future changes in climate related to climate change.

The approach to incorporating climate may be different from region to region, but the TMI could be calculated using historical data obtained from meteorological stations as well as incorporating future variations in project site location.

Future efforts should focus on creating tools that integrate climate information from meteorological stations and climate projections. These tools can be used to enhance pavement design methods, ultimately leading to the construction of more resilient structures to face climate and climate change.

The challenge for some countries will be to obtain climate data due to the limited availability of weather stations and historical data recording.

Finally, it is important to consider the road drainage system in pavement design, since effective road design is critical to the durability and sustainability of the pavement.